A comparative study of efficient initialization methods for the k-means clustering algorithm

► K-means is the most widely used partitional clustering algorithm. ► k-means is highly sensitive to the selection of the initial centers. ► We present an overview of k-means initialization methods (IMs). ► We then compare eight commonly used linear time IMs. ► We demonstrate that popular IMs often...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Expert systems with applications Ročník 40; číslo 1; s. 200 - 210
Hlavní autoři:	Celebi, M. Emre, Kingravi, Hassan A., Vela, Patricio A.
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Amsterdam Elsevier Ltd 01.01.2013 Elsevier
Témata:	Algorithmics. Computability. Computer arithmetics Algorithms Applied sciences Cluster analysis Cluster center initialization Clustering Clusters Collection Computational efficiency Computer science; control theory; systems Data processing. List processing. Character string processing Exact sciences and technology Expert systems k-means Memory organisation. Data processing Partitional clustering Software Statistical tests Sum of squared error criterion Theoretical computing k-means Sum of squared error criterion Partitional clustering Cluster center initialization Initialization Gradient Non parametric test Statistical analysis Cluster Linear time K means algorithm Recommendation Search algorithm Gradient descent Experimental result Efficiency Classification Descent method Database Linear complexity Alternative method Occupation time Time complexity
ISSN:	0957-4174, 1873-6793
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Abstract	► K-means is the most widely used partitional clustering algorithm. ► k-means is highly sensitive to the selection of the initial centers. ► We present an overview of k-means initialization methods (IMs). ► We then compare eight commonly used linear time IMs. ► We demonstrate that popular IMs often perform poorly. K-means is undoubtedly the most widely used partitional clustering algorithm. Unfortunately, due to its gradient descent nature, this algorithm is highly sensitive to the initial placement of the cluster centers. Numerous initialization methods have been proposed to address this problem. In this paper, we first present an overview of these methods with an emphasis on their computational efficiency. We then compare eight commonly used linear time complexity initialization methods on a large and diverse collection of data sets using various performance criteria. Finally, we analyze the experimental results using non-parametric statistical tests and provide recommendations for practitioners. We demonstrate that popular initialization methods often perform poorly and that there are in fact strong alternatives to these methods.
AbstractList	K-means is undoubtedly the most widely used partitional clustering algorithm. Unfortunately, due to its gradient descent nature, this algorithm is highly sensitive to the initial placement of the cluster centers. Numerous initialization methods have been proposed to address this problem. In this paper, we first present an overview of these methods with an emphasis on their computational efficiency. We then compare eight commonly used linear time complexity initialization methods on a large and diverse collection of data sets using various performance criteria. Finally, we analyze the experimental results using non-parametric statistical tests and provide recommendations for practitioners. We demonstrate that popular initialization methods often perform poorly and that there are in fact strong alternatives to these methods. ► K-means is the most widely used partitional clustering algorithm. ► k-means is highly sensitive to the selection of the initial centers. ► We present an overview of k-means initialization methods (IMs). ► We then compare eight commonly used linear time IMs. ► We demonstrate that popular IMs often perform poorly. K-means is undoubtedly the most widely used partitional clustering algorithm. Unfortunately, due to its gradient descent nature, this algorithm is highly sensitive to the initial placement of the cluster centers. Numerous initialization methods have been proposed to address this problem. In this paper, we first present an overview of these methods with an emphasis on their computational efficiency. We then compare eight commonly used linear time complexity initialization methods on a large and diverse collection of data sets using various performance criteria. Finally, we analyze the experimental results using non-parametric statistical tests and provide recommendations for practitioners. We demonstrate that popular initialization methods often perform poorly and that there are in fact strong alternatives to these methods.
Author	Celebi, M. Emre Kingravi, Hassan A. Vela, Patricio A.
Author_xml	– sequence: 1 givenname: M. Emre surname: Celebi fullname: Celebi, M. Emre email: ecelebi@lsus.edu organization: Department of Computer Science, Louisiana State University, Shreveport, LA, USA – sequence: 2 givenname: Hassan A. surname: Kingravi fullname: Kingravi, Hassan A. email: kingravi@gatech.edu organization: School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USA – sequence: 3 givenname: Patricio A. surname: Vela fullname: Vela, Patricio A. email: pvela@gatech.edu organization: School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USA
BackLink	http://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&idt=27095898$$DView record in Pascal Francis
BookMark	eNqNkT1rHDEQhkVwIGfHf8CVmkCaXY-kvZUW0hiTD4MhTdIZhJBGPl12pYukc3B-fXQ-p0lhUgmG5xlG73tKTmKKSMgFg54BGy-3PZZfpufAeA-yB85ekRVTUnSjnMQJWcG0lt3A5PCGnJayBWASQK7I3RW1admZbGp4QFrq3j3S5Cl6H2zAWGmIoQYzh9-NSJEuWDfJFepTpnWD9Ee3oImF2nlfKuYQ76mZ71MOdbO8Ja-9mQueP79n5Punj9-uv3S3Xz_fXF_ddnYYeW1njRPzkk-DVFY4a8HBBAYcU8LByN3kBa6Nt2vlYUQ1GtEmSlhmuJVKiDPy_rh3l9PPPZaql1AszrOJmPZFt78y4ELx_0DFMA3jxJls6Ltn1BRrZp9NtKHoXQ6LyY-ay5apmlTj1JGzOZWS0Wsb6lNYNZswawb60JHe6kNH-tCRBqlbR03l_6h_t78ofThK2CJ9CJh1OTRl0YWMtmqXwkv6H0-VrR0
CitedBy_id	crossref_primary_10_1007_s11128_021_03240_8 crossref_primary_10_1016_j_eswa_2024_123298 crossref_primary_10_1051_matecconf_201818903002 crossref_primary_10_3934_aci_2024004 crossref_primary_10_1007_s11634_022_00514_6 crossref_primary_10_1109_ACCESS_2023_3327640 crossref_primary_10_1016_j_eswa_2016_10_005 crossref_primary_10_1186_s40537_025_01119_4 crossref_primary_10_3233_IDT_170318 crossref_primary_10_2139_ssrn_5368329 crossref_primary_10_1002_sam_11416 crossref_primary_10_1016_j_margeo_2020_106332 crossref_primary_10_1109_ACCESS_2021_3066348 crossref_primary_10_1007_s42452_020_3157_6 crossref_primary_10_1016_j_asoc_2019_02_038 crossref_primary_10_1007_s10586_024_04305_w crossref_primary_10_7717_peerj_cs_234 crossref_primary_10_1016_j_oceaneng_2020_107329 crossref_primary_10_1016_j_petrol_2022_110599 crossref_primary_10_1016_j_neuroimage_2020_117464 crossref_primary_10_1109_TII_2020_2977980 crossref_primary_10_1016_j_conbuildmat_2022_128802 crossref_primary_10_1016_j_jksuci_2018_10_015 crossref_primary_10_1016_j_jobe_2022_105219 crossref_primary_10_1002_cpe_5969 crossref_primary_10_1016_j_rser_2019_109628 crossref_primary_10_1016_j_jobe_2022_105590 crossref_primary_10_1007_s10489_023_04580_x crossref_primary_10_1016_j_asoc_2022_109838 crossref_primary_10_3390_math13132138 crossref_primary_10_1007_s12553_023_00805_8 crossref_primary_10_1016_j_infrared_2020_103286 crossref_primary_10_1016_j_eswa_2020_114224 crossref_primary_10_1155_2016_4091323 crossref_primary_10_3390_e21101013 crossref_primary_10_1016_j_eswa_2023_120999 crossref_primary_10_1016_j_fuel_2020_117084 crossref_primary_10_1088_1742_6596_1235_1_012015 crossref_primary_10_1016_j_jqsrt_2020_107003 crossref_primary_10_1007_s10489_020_01677_5 crossref_primary_10_1016_j_chaos_2021_111494 crossref_primary_10_1061_JPCFEV_CFENG_4238 crossref_primary_10_1177_10812865231174219 crossref_primary_10_1061__ASCE_CO_1943_7862_0002406 crossref_primary_10_1080_03081060_2025_2457055 crossref_primary_10_3390_en16104129 crossref_primary_10_1109_ACCESS_2022_3219457 crossref_primary_10_1007_s10846_021_01481_4 crossref_primary_10_1109_TKDE_2015_2453162 crossref_primary_10_1007_s00500_020_04988_4 crossref_primary_10_1007_s10660_021_09458_z crossref_primary_10_1186_s13640_018_0260_3 crossref_primary_10_1007_s11063_017_9755_7 crossref_primary_10_3390_rs16173230 crossref_primary_10_1007_s11042_018_7100_4 crossref_primary_10_3390_smartcities8050145 crossref_primary_10_1016_j_knosys_2016_11_020 crossref_primary_10_3389_fbuil_2023_1336280 crossref_primary_10_1016_j_engappai_2021_104532 crossref_primary_10_1007_s10559_023_00587_x crossref_primary_10_1007_s00170_019_04849_x crossref_primary_10_2478_bsrj_2019_019 crossref_primary_10_1109_ACCESS_2021_3136435 crossref_primary_10_1007_s00521_022_07554_1 crossref_primary_10_1016_j_trd_2024_104456 crossref_primary_10_1007_s41066_023_00406_w crossref_primary_10_1007_s11042_019_7652_y crossref_primary_10_1016_j_energy_2018_05_029 crossref_primary_10_3390_s24165180 crossref_primary_10_1177_1847980416673784 crossref_primary_10_34133_cbsystems_0381 crossref_primary_10_1007_s40964_025_01199_x crossref_primary_10_1038_s41598_019_38845_8 crossref_primary_10_1109_ACCESS_2023_3272403 crossref_primary_10_1186_s12889_024_19369_x crossref_primary_10_3390_electronics14091723 crossref_primary_10_1016_j_rser_2024_114647 crossref_primary_10_3390_land14061228 crossref_primary_10_1016_j_compag_2025_110841 crossref_primary_10_1007_s00217_018_3137_x crossref_primary_10_3390_a16070349 crossref_primary_10_26634_jcc_4_2_14382 crossref_primary_10_1016_j_tbs_2024_100808 crossref_primary_10_3390_e22080902 crossref_primary_10_1016_j_patrec_2019_02_011 crossref_primary_10_1088_1742_6596_1546_1_012085 crossref_primary_10_1038_s41524_021_00565_x crossref_primary_10_1109_JIOT_2022_3166455 crossref_primary_10_1016_j_patcog_2025_111654 crossref_primary_10_1007_s12530_022_09447_z crossref_primary_10_1109_TPDS_2014_2306193 crossref_primary_10_1007_s11053_025_10504_y crossref_primary_10_1016_j_ejor_2017_02_013 crossref_primary_10_1007_s12046_019_1166_1 crossref_primary_10_1109_ACCESS_2020_2978404 crossref_primary_10_3390_ijerph16234748 crossref_primary_10_1002_ldr_4211 crossref_primary_10_1016_j_engappai_2018_05_004 crossref_primary_10_1109_ACCESS_2022_3179803 crossref_primary_10_1109_JBHI_2016_2633403 crossref_primary_10_3390_e24010028 crossref_primary_10_3390_rs13234922 crossref_primary_10_1016_j_imu_2019_01_001 crossref_primary_10_1371_journal_pone_0267009 crossref_primary_10_3389_frvir_2022_782564 crossref_primary_10_1016_j_enganabound_2024_105919 crossref_primary_10_1016_j_compchemeng_2020_106810 crossref_primary_10_1049_stg2_70009 crossref_primary_10_1186_s12909_023_04936_4 crossref_primary_10_1016_j_ijcchd_2024_100524 crossref_primary_10_1016_j_future_2017_10_043 crossref_primary_10_1016_j_neucom_2017_04_073 crossref_primary_10_1007_s10044_020_00887_4 crossref_primary_10_1155_2014_380480 crossref_primary_10_3390_app14167379 crossref_primary_10_1080_17517575_2025_2454004 crossref_primary_10_1016_j_gexplo_2024_107385 crossref_primary_10_1177_09544070231205063 crossref_primary_10_1016_j_jpowsour_2019_227259 crossref_primary_10_3390_e25030510 crossref_primary_10_1109_ACCESS_2022_3206371 crossref_primary_10_1109_TAP_2021_3118710 crossref_primary_10_1017_S1368980014003243 crossref_primary_10_1016_j_scs_2017_08_002 crossref_primary_10_3390_foods11203270 crossref_primary_10_1016_j_diabres_2024_111803 crossref_primary_10_3938_jkps_74_298 crossref_primary_10_1007_s42102_024_00122_2 crossref_primary_10_1007_s11047_016_9542_9 crossref_primary_10_1007_s10586_020_03120_3 crossref_primary_10_1016_j_infrared_2024_105403 crossref_primary_10_1007_s10044_021_01045_0 crossref_primary_10_1016_j_rser_2021_111162 crossref_primary_10_1007_s12652_022_04428_1 crossref_primary_10_4018_ijdwm_2014070101 crossref_primary_10_1007_s00521_022_07992_x crossref_primary_10_1007_s10260_015_0345_4 crossref_primary_10_3390_e21121145 crossref_primary_10_1007_s11634_019_00356_9 crossref_primary_10_1016_j_imu_2016_01_002 crossref_primary_10_1007_s00521_020_04747_4 crossref_primary_10_1109_TIT_2020_3008058 crossref_primary_10_1016_j_patcog_2020_107713 crossref_primary_10_1007_s00146_025_02430_7 crossref_primary_10_1108_K_06_2023_1044 crossref_primary_10_2478_amns_2023_2_00486 crossref_primary_10_1061_JTEPBS_TEENG_8277 crossref_primary_10_1109_JSYST_2019_2918234 crossref_primary_10_1109_TVCG_2019_2934256 crossref_primary_10_1016_j_lungcan_2023_107308 crossref_primary_10_1109_TEM_2020_3018912 crossref_primary_10_1186_s13007_024_01297_x crossref_primary_10_1016_j_knosys_2019_02_035 crossref_primary_10_1080_27684830_2022_2073662 crossref_primary_10_3390_technologies13030123 crossref_primary_10_1109_TAP_2018_2876192 crossref_primary_10_1007_s00484_019_01699_w crossref_primary_10_1007_s10044_017_0673_0 crossref_primary_10_3390_buildings13040962 crossref_primary_10_1109_TCSI_2013_2239098 crossref_primary_10_3389_fncom_2023_1243092 crossref_primary_10_1016_j_enconman_2016_04_009 crossref_primary_10_1002_cpe_7185 crossref_primary_10_1007_s10462_023_10406_6 crossref_primary_10_1134_S0965542518040140 crossref_primary_10_1016_j_envres_2020_110633 crossref_primary_10_1177_0040517520957401 crossref_primary_10_1016_j_tra_2021_10_001 crossref_primary_10_1016_j_solener_2014_10_002 crossref_primary_10_1016_S1005_8885_16_60029_8 crossref_primary_10_1371_journal_pone_0162259 crossref_primary_10_1002_apj_2599 crossref_primary_10_1016_j_jprocont_2019_08_001 crossref_primary_10_1007_s11042_023_14388_z crossref_primary_10_1007_s11334_020_00372_5 crossref_primary_10_1016_j_eswa_2014_12_027 crossref_primary_10_1007_s11280_021_00945_9 crossref_primary_10_1007_s00521_020_05471_9 crossref_primary_10_2139_ssrn_3805831 crossref_primary_10_1155_2016_4606384 crossref_primary_10_7717_peerj_cs_1302 crossref_primary_10_1088_1757_899X_598_1_012116 crossref_primary_10_1186_s10033_018_0202_0 crossref_primary_10_1007_s11042_018_5649_6 crossref_primary_10_1111_1365_2664_14577 crossref_primary_10_1016_j_asoc_2019_03_013 crossref_primary_10_1016_j_ebiom_2019_01_028 crossref_primary_10_1016_j_ctro_2019_12_004 crossref_primary_10_1016_j_cageo_2022_105232 crossref_primary_10_1016_j_engappai_2025_110370 crossref_primary_10_1016_j_procs_2021_01_292 crossref_primary_10_1016_j_petrol_2018_10_005 crossref_primary_10_1007_s10489_021_02405_3 crossref_primary_10_1016_j_patcog_2014_01_015 crossref_primary_10_3390_molecules28196886 crossref_primary_10_1038_s41598_021_85159_9 crossref_primary_10_1016_j_ijpe_2019_05_020 crossref_primary_10_1109_TVCG_2023_3330262 crossref_primary_10_3390_agriculture12030428 crossref_primary_10_1016_j_trd_2018_07_012 crossref_primary_10_1007_s10489_024_05636_2 crossref_primary_10_1108_JAMR_07_2021_0242 crossref_primary_10_1016_j_infsof_2020_106501 crossref_primary_10_1109_TBME_2013_2297622 crossref_primary_10_3390_soc15090242 crossref_primary_10_1049_iet_ipr_2020_0709 crossref_primary_10_1109_TIP_2021_3135470 crossref_primary_10_1016_j_neucom_2016_01_105 crossref_primary_10_3390_land12122156 crossref_primary_10_1016_j_neucom_2021_02_059 crossref_primary_10_1016_j_knosys_2019_05_019 crossref_primary_10_1109_JBHI_2018_2823499 crossref_primary_10_1016_j_measen_2024_101237 crossref_primary_10_1177_21582440211027574 crossref_primary_10_1186_s12888_016_0966_7 crossref_primary_10_1016_j_childyouth_2025_108559 crossref_primary_10_1093_nar_gkab015 crossref_primary_10_1016_j_compbiomed_2017_10_014 crossref_primary_10_1109_JSEN_2025_3528956 crossref_primary_10_1016_j_segan_2023_101117 crossref_primary_10_1007_s10291_025_01937_2 crossref_primary_10_1007_s11442_023_2082_1 crossref_primary_10_1016_j_neunet_2023_09_033 crossref_primary_10_1007_s00521_016_2534_y crossref_primary_10_1016_j_eswa_2015_05_014 crossref_primary_10_1088_1757_899X_928_3_032059 crossref_primary_10_1016_j_patcog_2016_05_005 crossref_primary_10_1016_j_cma_2025_118120 crossref_primary_10_1007_s10489_025_06367_8 crossref_primary_10_1111_exsy_12491 crossref_primary_10_1007_s12652_020_02762_w crossref_primary_10_1002_ente_202200784 crossref_primary_10_1016_j_eswa_2022_117927 crossref_primary_10_1007_s11760_024_03621_3 crossref_primary_10_1007_s00500_020_05392_8 crossref_primary_10_1109_ACCESS_2023_3243195 crossref_primary_10_1016_j_egypro_2017_03_705 crossref_primary_10_1007_s10111_022_00706_2 crossref_primary_10_1016_j_multra_2022_100049 crossref_primary_10_1016_j_eswa_2020_114035 crossref_primary_10_22630_MIBE_2023_24_4_19 crossref_primary_10_1109_JSEN_2017_2755547 crossref_primary_10_1016_j_ijhydene_2023_10_204 crossref_primary_10_3390_e21020100 crossref_primary_10_1007_s13369_024_09628_9 crossref_primary_10_1371_journal_pone_0127125 crossref_primary_10_1109_TPAMI_2023_3237667 crossref_primary_10_3390_rs11242994 crossref_primary_10_1016_j_engappai_2017_11_002 crossref_primary_10_1002_dac_3387 crossref_primary_10_1016_j_jhydrol_2021_126449 crossref_primary_10_1109_TITS_2021_3076607 crossref_primary_10_1002_cai2_68 crossref_primary_10_1007_s11554_018_0814_8 crossref_primary_10_3390_rs13051002 crossref_primary_10_1016_j_eswa_2021_115558 crossref_primary_10_1007_s11042_018_6324_7 crossref_primary_10_1049_iet_rpg_2016_0240 crossref_primary_10_1111_exsy_13248 crossref_primary_10_1016_j_cor_2022_105958 crossref_primary_10_1080_08911762_2013_878428 crossref_primary_10_1109_JIOT_2020_2981774 crossref_primary_10_1016_j_physa_2020_125567 crossref_primary_10_1002_eng2_70104 crossref_primary_10_1109_ACCESS_2020_2973216 crossref_primary_10_1016_j_apenergy_2024_123007 crossref_primary_10_1016_j_jii_2025_100788 crossref_primary_10_1007_s10346_020_01473_9 crossref_primary_10_1109_ACCESS_2019_2907885 crossref_primary_10_1109_TII_2021_3104111 crossref_primary_10_1016_j_flowmeasinst_2025_102852 crossref_primary_10_1016_j_ndteint_2025_103362 crossref_primary_10_1016_j_eswa_2018_03_052 crossref_primary_10_1016_j_patcog_2019_04_014 crossref_primary_10_1007_s10115_016_0946_8 crossref_primary_10_1177_20414196241281069 crossref_primary_10_1016_j_ebiom_2019_04_055 crossref_primary_10_3390_s151129056 crossref_primary_10_1016_j_jprocont_2023_103080 crossref_primary_10_1002_jbio_201700056 crossref_primary_10_1016_j_compbiomed_2019_01_017 crossref_primary_10_1155_2015_615740 crossref_primary_10_1109_ACCESS_2023_3341156 crossref_primary_10_1016_j_patrec_2014_11_017 crossref_primary_10_1016_j_patrec_2018_11_017 crossref_primary_10_1088_1742_6596_1641_1_012081 crossref_primary_10_4028_p_68Ypy6 crossref_primary_10_1007_s10994_021_06021_7 crossref_primary_10_1007_s10044_021_00977_x crossref_primary_10_1007_s10115_024_02066_x crossref_primary_10_1016_j_patcog_2018_02_015 crossref_primary_10_1007_s10489_020_02067_7 crossref_primary_10_1016_j_solener_2014_01_021 crossref_primary_10_1007_s00477_021_01981_7 crossref_primary_10_1109_TKDE_2022_3155450 crossref_primary_10_1093_clinchem_hvab239 crossref_primary_10_1038_s41598_020_59553_8 crossref_primary_10_1007_s00521_019_04467_4 crossref_primary_10_1016_j_ins_2022_11_139 crossref_primary_10_1108_IMDS_01_2017_0029 crossref_primary_10_1016_j_ymssp_2020_106693 crossref_primary_10_1016_j_knosys_2020_106268 crossref_primary_10_1109_ACCESS_2023_3239212 crossref_primary_10_1007_s00267_018_1101_y crossref_primary_10_1109_TCSS_2023_3251319 crossref_primary_10_4018_IJISMD_2019010103 crossref_primary_10_1016_j_cor_2024_106548 crossref_primary_10_3390_a16040215 crossref_primary_10_1007_s13198_021_01424_0 crossref_primary_10_3390_e21050487 crossref_primary_10_1016_j_procs_2020_08_022 crossref_primary_10_1007_s11042_018_6966_5 crossref_primary_10_1136_bmjmed_2021_000070 crossref_primary_10_1016_j_apenergy_2021_118387 crossref_primary_10_1179_1743131X13Y_0000000059 crossref_primary_10_1016_j_eswa_2017_01_024 crossref_primary_10_1016_j_future_2017_12_012 crossref_primary_10_3390_rs6098424 crossref_primary_10_1109_TVCG_2023_3326601 crossref_primary_10_22581_muet1982_2103_16 crossref_primary_10_18697_ajfand_96_19775 crossref_primary_10_3390_rs17020187 crossref_primary_10_1016_j_marpetgeo_2022_105734 crossref_primary_10_1007_s13369_023_08007_0 crossref_primary_10_3390_su14052744 crossref_primary_10_1038_s41598_024_84082_z crossref_primary_10_1007_s10772_018_9523_8 crossref_primary_10_1088_1742_6596_1992_4_042060 crossref_primary_10_1109_LAWP_2021_3060190 crossref_primary_10_1007_s43684_023_00055_5 crossref_primary_10_3390_en11081975 crossref_primary_10_1016_j_procs_2017_08_210 crossref_primary_10_3390_en12071248 crossref_primary_10_1109_ACCESS_2022_3230935 crossref_primary_10_1109_TIM_2025_3582311 crossref_primary_10_1007_s00500_019_04216_8 crossref_primary_10_1016_j_patrec_2017_10_031 crossref_primary_10_1177_0165551518816302 crossref_primary_10_1016_j_asoc_2016_07_022 crossref_primary_10_1038_s41598_018_21988_5 crossref_primary_10_1007_s12652_018_0697_3 crossref_primary_10_3390_info9040101 crossref_primary_10_1088_1742_6596_930_1_012021 crossref_primary_10_4018_IJDWM_316142 crossref_primary_10_1007_s10514_025_10199_3 crossref_primary_10_3390_jimaging11060173 crossref_primary_10_1051_0004_6361_202555620 crossref_primary_10_1016_j_eswa_2018_09_006 crossref_primary_10_1155_2019_9610212 crossref_primary_10_1007_s10044_024_01228_5 crossref_primary_10_4018_IJIRR_289954 crossref_primary_10_1145_3689036 crossref_primary_10_3390_rs12223708 crossref_primary_10_3390_w12092433 crossref_primary_10_1016_j_aei_2024_102799 crossref_primary_10_1177_2374289519873088 crossref_primary_10_1109_JIOT_2019_2935010 crossref_primary_10_20517_jmi_2024_85 crossref_primary_10_3390_math9121423 crossref_primary_10_1177_1729881420912142 crossref_primary_10_7717_peerj_cs_2286 crossref_primary_10_1088_1361_6501_acdf76 crossref_primary_10_1117_1_JEI_33_5_053052 crossref_primary_10_3390_fi15060195 crossref_primary_10_1049_iet_rpg_2016_0982 crossref_primary_10_1007_s00357_018_9296_4 crossref_primary_10_1016_j_cag_2021_10_015 crossref_primary_10_1016_j_diin_2017_05_001 crossref_primary_10_3390_metrology5030048 crossref_primary_10_3233_JIFS_18146 crossref_primary_10_1051_e3sconf_201912525001 crossref_primary_10_1109_ACCESS_2024_3497977 crossref_primary_10_1186_s40064_016_3329_4 crossref_primary_10_1109_ACCESS_2025_3586081 crossref_primary_10_1007_s13042_020_01190_8 crossref_primary_10_1016_j_procs_2016_09_080 crossref_primary_10_1109_TNNLS_2021_3071083 crossref_primary_10_1016_j_trgeo_2023_101053 crossref_primary_10_1007_s11634_022_00510_w crossref_primary_10_1016_j_eswa_2017_08_050 crossref_primary_10_1109_ACCESS_2021_3085005 crossref_primary_10_1002_joc_8396 crossref_primary_10_1007_s10758_020_09468_0 crossref_primary_10_1007_s10278_022_00653_4 crossref_primary_10_1007_s00607_023_01249_8 crossref_primary_10_3390_app11104406 crossref_primary_10_3390_su11174635 crossref_primary_10_1016_j_engfailanal_2022_106696 crossref_primary_10_1155_2022_3958423 crossref_primary_10_1109_TAP_2022_3143137 crossref_primary_10_1016_j_cageo_2013_04_028 crossref_primary_10_1007_s00778_021_00716_y crossref_primary_10_1007_s00357_020_09372_3 crossref_primary_10_1109_TFUZZ_2018_2889014 crossref_primary_10_1002_ecs2_4811 crossref_primary_10_1002_sce_21743 crossref_primary_10_1016_j_petrol_2021_109335 crossref_primary_10_1109_ACCESS_2019_2927757 crossref_primary_10_1002_advs_202206982 crossref_primary_10_1007_s40031_014_0106_z crossref_primary_10_1016_j_knosys_2020_106482 crossref_primary_10_1049_joe_2017_0090 crossref_primary_10_1049_rpg2_12875 crossref_primary_10_1016_j_ecolind_2017_12_022 crossref_primary_10_4018_ijisss_2015040105 crossref_primary_10_1002_ijfe_2344 crossref_primary_10_1016_j_ins_2017_07_036 crossref_primary_10_1007_s11760_019_01469_6 crossref_primary_10_32604_cmc_2024_060090 crossref_primary_10_2478_acss_2023_0001 crossref_primary_10_1007_s41060_023_00452_2 crossref_primary_10_1109_TII_2019_2933675 crossref_primary_10_1109_TVCG_2021_3114839 crossref_primary_10_3390_a16120572 crossref_primary_10_1177_23998083241293833 crossref_primary_10_1007_s00500_018_3565_3 crossref_primary_10_1080_0305764X_2022_2061914 crossref_primary_10_1016_j_cag_2017_01_001 crossref_primary_10_1016_j_neucom_2025_130227 crossref_primary_10_1111_jedm_12208 crossref_primary_10_1016_j_compmedimag_2019_101643 crossref_primary_10_1109_TAP_2023_3304748 crossref_primary_10_1016_j_cie_2020_106290 crossref_primary_10_1109_ACCESS_2024_3350442 crossref_primary_10_1016_j_eswa_2017_06_022 crossref_primary_10_1109_ACCESS_2021_3057113 crossref_primary_10_1016_j_eswa_2015_02_006 crossref_primary_10_24818_EA_2023_62_12 crossref_primary_10_3390_su14148473 crossref_primary_10_1016_j_bspc_2020_102244 crossref_primary_10_1016_j_knosys_2023_111071 crossref_primary_10_3390_app12157378 crossref_primary_10_3390_en12010102 crossref_primary_10_1007_s11227_021_03958_3 crossref_primary_10_1016_j_bpj_2024_03_023 crossref_primary_10_1002_ett_3647 crossref_primary_10_1287_opre_2023_0398 crossref_primary_10_1111_coin_12297 crossref_primary_10_1016_j_compmedimag_2015_02_011 crossref_primary_10_1080_2157930X_2019_1685792 crossref_primary_10_1016_j_ijrmms_2024_105879 crossref_primary_10_1109_JSYST_2014_2313671 crossref_primary_10_3390_electronics10222786 crossref_primary_10_1088_1757_899X_336_1_012017 crossref_primary_10_1109_LRA_2023_3264753 crossref_primary_10_3390_en17143573 crossref_primary_10_1177_1550147717728627 crossref_primary_10_1016_j_asoc_2025_113559 crossref_primary_10_1016_j_eswa_2022_118656 crossref_primary_10_1049_joe_2019_0938 crossref_primary_10_1016_j_apacoust_2017_08_008 crossref_primary_10_1016_j_matchar_2023_113607 crossref_primary_10_1016_j_suscom_2021_100561 crossref_primary_10_1109_TKDE_2020_3021649 crossref_primary_10_3390_math8071070 crossref_primary_10_2166_ws_2020_054 crossref_primary_10_3390_s18082713 crossref_primary_10_1177_1094428117752467 crossref_primary_10_1002_cpe_4109 crossref_primary_10_1007_s00170_024_14601_9 crossref_primary_10_1007_s11760_019_01417_4 crossref_primary_10_1016_j_foreco_2023_121061 crossref_primary_10_1007_s13278_019_0559_9 crossref_primary_10_3233_IDA_192791 crossref_primary_10_1007_s13369_017_2761_2 crossref_primary_10_3934_bdia_2016_1_93 crossref_primary_10_1038_s41598_025_87672_7 crossref_primary_10_1007_s11042_024_18175_2 crossref_primary_10_1186_s13638_020_01853_8 crossref_primary_10_1007_s10044_025_01463_4 crossref_primary_10_1016_j_artmed_2019_05_002 crossref_primary_10_32604_cmc_2025_057693 crossref_primary_10_1016_j_asoc_2021_107899 crossref_primary_10_1016_j_eswa_2020_113435 crossref_primary_10_1155_2014_761468 crossref_primary_10_1109_JBHI_2018_2845939 crossref_primary_10_1016_j_chemolab_2021_104408 crossref_primary_10_1080_00207543_2024_2341415 crossref_primary_10_1016_j_buildenv_2022_109473 crossref_primary_10_1007_s10618_014_0395_5 crossref_primary_10_3847_1538_4357_ade805 crossref_primary_10_1016_j_eswa_2014_07_014 crossref_primary_10_4018_IJHCITP_2018010101 crossref_primary_10_1016_j_bbcan_2021_188588 crossref_primary_10_3390_s16122047 crossref_primary_10_1109_ACCESS_2021_3067060 crossref_primary_10_1016_j_scs_2019_101958 crossref_primary_10_1061__ASCE_PS_1949_1204_0000425 crossref_primary_10_3390_ijgi6120392 crossref_primary_10_1109_TVT_2019_2924669 crossref_primary_10_1007_JHEP08_2021_170 crossref_primary_10_1016_j_datak_2018_04_001 crossref_primary_10_1016_j_compeleceng_2018_04_023 crossref_primary_10_1016_j_eswa_2016_03_008 crossref_primary_10_1016_j_asoc_2017_11_038 crossref_primary_10_1002_widm_1330 crossref_primary_10_1080_1206212X_2020_1735035 crossref_primary_10_1016_j_imu_2019_100239 crossref_primary_10_1016_j_asr_2024_09_013 crossref_primary_10_1016_j_neucom_2024_128101 crossref_primary_10_1016_j_neucom_2019_05_056 crossref_primary_10_1007_s13748_023_00304_x crossref_primary_10_1016_j_eswa_2018_07_029 crossref_primary_10_1007_s12083_018_0668_7 crossref_primary_10_1007_s12530_018_9235_y crossref_primary_10_1016_j_ress_2021_108230 crossref_primary_10_1109_TEVC_2022_3144134 crossref_primary_10_1155_2018_6360741 crossref_primary_10_1002_cae_22452 crossref_primary_10_1145_2873064 crossref_primary_10_7603_s40632_016_0010_6 crossref_primary_10_1088_1361_6501_aca2ce crossref_primary_10_2478_jses_2023_0007 crossref_primary_10_1007_s12065_022_00720_3 crossref_primary_10_1155_2014_847608 crossref_primary_10_1109_TNSM_2017_2785660 crossref_primary_10_3390_app9061215 crossref_primary_10_1016_j_patcog_2019_106969 crossref_primary_10_1002_jrs_70023 crossref_primary_10_1016_j_fuel_2025_135433 crossref_primary_10_1186_s11671_016_1254_7 crossref_primary_10_3233_HIS_190277 crossref_primary_10_1007_s11554_019_00914_6 crossref_primary_10_1016_j_inffus_2024_102411 crossref_primary_10_1061_JPCFEV_CFENG_5008 crossref_primary_10_1016_j_enbuild_2021_110862 crossref_primary_10_3390_su16219244 crossref_primary_10_1080_00051144_2023_2253064 crossref_primary_10_1007_s11042_015_2790_3 crossref_primary_10_3390_buildings13030666 crossref_primary_10_3390_economies13050145 crossref_primary_10_1016_j_rser_2023_114191 crossref_primary_10_3390_a15040117 crossref_primary_10_1007_s10201_015_0454_7 crossref_primary_10_1016_j_scitotenv_2024_170765 crossref_primary_10_1186_s40494_024_01195_4 crossref_primary_10_1007_s12065_020_00562_x crossref_primary_10_3390_agriengineering6040250 crossref_primary_10_1007_s12065_023_00863_x crossref_primary_10_1016_j_jqsrt_2018_04_014 crossref_primary_10_1080_21681376_2022_2132180 crossref_primary_10_3390_app9122432 crossref_primary_10_1261_rna_074427_119 crossref_primary_10_3390_app112311524 crossref_primary_10_3390_rs9060532 crossref_primary_10_1016_j_ecolmodel_2019_03_021 crossref_primary_10_1186_s40959_024_00276_4 crossref_primary_10_1016_j_knosys_2021_107230 crossref_primary_10_1007_s00500_018_3289_4 crossref_primary_10_1155_2022_7518422 crossref_primary_10_1109_JSEN_2021_3058717 crossref_primary_10_1109_TQE_2022_3185505 crossref_primary_10_3390_a15060191 crossref_primary_10_1016_j_bbr_2021_113223 crossref_primary_10_1109_TKDE_2020_3002926 crossref_primary_10_1007_s11042_023_18067_x crossref_primary_10_1115_1_4069681
Cites_doi	10.1109/72.478389 10.1016/j.eswa.2008.11.041 10.1016/0167-8655(93)90058-L 10.1145/331499.331504 10.1016/j.eswa.2008.06.093 10.1002/bs.3830120210 10.1016/j.patrec.2009.09.011 10.1109/TKDE.2004.25 10.1109/TCOM.1985.1096214 10.1016/S0031-3203(03)00190-0 10.1109/TIT.1982.1056489 10.1198/jcgs.2009.08054 10.1007/s00500-008-0392-y 10.1007/BF01908075 10.1007/s10994-009-5103-0 10.1016/j.imavis.2010.10.002 10.1016/j.tcs.2010.05.034 10.1016/j.patrec.2007.01.001 10.1145/1557019.1557115 10.1109/TSMCB.2003.816993 10.1109/TPAMI.2010.88 10.1016/S0377-2217(77)81005-9 10.1016/j.patrec.2009.04.013 10.3233/IDA-2007-11402 10.1016/0167-8655(95)00119-0 10.1007/978-3-540-48247-5_62 10.1109/72.761722 10.1109/TSMC.1977.4309789 10.1109/TPAMI.2002.1017616 10.1109/TPAMI.1984.4767478 10.1109/TVLSI.2009.2017543 10.1109/83.210871 10.1007/BF02287921 10.1071/BT9660127 10.1016/S0031-3203(02)00060-2 10.1016/j.camwa.2009.04.017 10.1016/j.jmva.2006.11.013 10.1016/0304-3975(85)90224-5 10.1016/S0167-8655(99)00069-0 10.1145/1830483.1830573 10.1080/01621459.1937.10503522 10.1137/1.9781611972801.12 10.1049/el:19901037 10.1109/97.329844 10.1145/335191.335388 10.1109/TCOM.1980.1094577 10.1007/BF01897163 10.1016/j.patrec.2007.12.009 10.4304/jetwi.4.1.51-59 10.1080/03610928008827904 10.1093/comjnl/10.3.271 10.1145/272991.272995 10.1007/BF02293907
ContentType	Journal Article
Copyright	2012 Elsevier Ltd 2014 INIST-CNRS
Copyright_xml	– notice: 2012 Elsevier Ltd – notice: 2014 INIST-CNRS
DBID	AAYXX CITATION IQODW 7SC 8FD JQ2 L7M L~C L~D
DOI	10.1016/j.eswa.2012.07.021
DatabaseName	CrossRef Pascal-Francis Computer and Information Systems Abstracts Technology Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional
DatabaseTitle	CrossRef Computer and Information Systems Abstracts Technology Research Database Computer and Information Systems Abstracts – Academic Advanced Technologies Database with Aerospace ProQuest Computer Science Collection Computer and Information Systems Abstracts Professional
DatabaseTitleList	Computer and Information Systems Abstracts Computer and Information Systems Abstracts
DeliveryMethod	fulltext_linktorsrc
Discipline	Computer Science Applied Sciences
EISSN	1873-6793
EndPage	210
ExternalDocumentID	27095898 10_1016_j_eswa_2012_07_021 S0957417412008767
GroupedDBID	--K --M .DC .~1 0R~ 13V 1B1 1RT 1~. 1~5 29G 4.4 457 4G. 5GY 5VS 7-5 71M 8P~ 9JN 9JO AAAKF AAAKG AABNK AACTN AAEDT AAEDW AAIAV AAIKJ AAKOC AALRI AAOAW AAQFI AAQXK AARIN AAXUO AAYFN ABBOA ABFNM ABKBG ABMAC ABMVD ABUCO ABXDB ABYKQ ACDAQ ACGFS ACHRH ACNNM ACNTT ACRLP ACZNC ADBBV ADEZE ADJOM ADMUD ADTZH AEBSH AECPX AEKER AENEX AFKWA AFTJW AGHFR AGJBL AGUBO AGUMN AGYEJ AHHHB AHJVU AHZHX AIALX AIEXJ AIKHN AITUG AJBFU AJOXV ALEQD ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AOUOD APLSM ASPBG AVWKF AXJTR AZFZN BJAXD BKOJK BLXMC BNSAS CS3 DU5 EBS EFJIC EFLBG EJD EO8 EO9 EP2 EP3 F5P FDB FEDTE FGOYB FIRID FNPLU FYGXN G-2 G-Q GBLVA GBOLZ HAMUX HLZ HVGLF HZ~ IHE J1W JJJVA KOM LG9 LY1 LY7 M41 MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. PQQKQ Q38 R2- RIG ROL RPZ SBC SDF SDG SDP SDS SES SET SEW SPC SPCBC SSB SSD SSL SST SSV SSZ T5K TN5 WUQ XPP ZMT ~G- 9DU AATTM AAXKI AAYWO AAYXX ABJNI ABUFD ABWVN ACLOT ACRPL ACVFH ADCNI ADNMO AEIPS AEUPX AFJKZ AFPUW AGQPQ AIGII AIIUN AKBMS AKRWK AKYEP ANKPU APXCP CITATION EFKBS ~HD 08R AALMO AAPBV ABPIF ABPTK ADALY IPNFZ IQODW PQEST 7SC 8FD JQ2 L7M L~C L~D
ID	FETCH-LOGICAL-c462t-41691f729478c3dcc0d090a0d183d062d9f3e5afc58f06e86a39f383c1a2c7833
ISICitedReferencesCount	808
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000309378200019&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	0957-4174
IngestDate	Sun Nov 09 10:38:26 EST 2025 Sun Nov 09 13:58:30 EST 2025 Fri Nov 25 01:07:34 EST 2022 Sat Nov 29 04:44:33 EST 2025 Tue Nov 18 22:25:29 EST 2025 Fri Feb 23 02:26:28 EST 2024
IsPeerReviewed	true
IsScholarly	true
Issue	1
Keywords	k-means Sum of squared error criterion Partitional clustering Cluster center initialization Initialization Gradient Non parametric test Statistical analysis Cluster Linear time K means algorithm Recommendation Search algorithm Gradient descent Experimental result Efficiency Classification Descent method Database Linear complexity Alternative method Occupation time Time complexity
Language	English
License	https://www.elsevier.com/tdm/userlicense/1.0 CC BY 4.0
LinkModel	OpenURL
MergedId	FETCHMERGED-LOGICAL-c462t-41691f729478c3dcc0d090a0d183d062d9f3e5afc58f06e86a39f383c1a2c7833
Notes	ObjectType-Article-2 SourceType-Scholarly Journals-1 ObjectType-Feature-1 content type line 23 ObjectType-Article-1 ObjectType-Feature-2
PQID	1349469217
PQPubID	23500
PageCount	11
ParticipantIDs	proquest_miscellaneous_1701023823 proquest_miscellaneous_1349469217 pascalfrancis_primary_27095898 crossref_citationtrail_10_1016_j_eswa_2012_07_021 crossref_primary_10_1016_j_eswa_2012_07_021 elsevier_sciencedirect_doi_10_1016_j_eswa_2012_07_021
PublicationCentury	2000
PublicationDate	January 2013 2013-1-00 2013 20130101
PublicationDateYYYYMMDD	2013-01-01
PublicationDate_xml	– month: 01 year: 2013 text: January 2013
PublicationDecade	2010
PublicationPlace	Amsterdam
PublicationPlace_xml	– name: Amsterdam
PublicationTitle	Expert systems with applications
PublicationYear	2013
Publisher	Elsevier Ltd Elsevier
Publisher_xml	– name: Elsevier Ltd – name: Elsevier
References	Chen, Song, Bai, Lin, Chang (b0100) 2011; 33 Hartigan, Wong (b0155) 1979; 28 Milligan, Cooper (b0290) 1988; 5 Pizzuti, C., Talia, D., & Vonella, G. (1999). A divisive initialisation method for clustering algorithms, In Chen, Chien (b0095) 2010; 18 (pp. 1027–1035). (pp. 297–302). (pp. 484–491). Anderberg (b0030) 1973 Cao, Liang, Jiang (b0085) 2009; 58 MacQueen, J. (1967). Some methods for classification and analysis of multivariate observations, In Lance, Williams (b0225) 1967; 10 Likas, Vlassis, Verbeek (b0230) 2003; 36 Tou, Gonzales (b0355) 1974 Aloise, Deshpande, Hansen, Popat (b0020) 2009; 75 Meilă (b0280) 2007; 98 (pp. 877–885). Maitra, Melnykov (b0265) 2010; 19 Voss, T., Hansen, N., & Igel, C. (2010). Improved step size adaptation for the MO-CMA-ES, In Zhang, Leung (b0380) 2003; 33 Jain (b0195) 2010; 31 (pp. 281–297). Huang, Harris (b0175) 1993; 2 Babu, Murty (b0045) 1993; 14 Norušis (b0295) 2011 Celebi (b0090) 2011; 29 Babu, Murty (b0050) 1994; 25 Tarsitano (b0350) 2003; 36 (pp. 32–39). Späth (b0340) 1977; 1 2’: Merging distance and density based clustering, In 13-21. Arthur, D., & Vassilvitskii, S. (2007). k-means++: The advantages of careful seeding, In Iman, Davenport (b0190) 1980; 9 (pp. 74–76). + , . Huang, Chen (b0170) 1990; 26 Daniel (b0110) 2000 Lloyd (b0240) 1982; 28 Lu, Tang, Tang, Yang (b0250) 2008; 29 1 Al-Daoud, M. (2005). A new algorithm for cluster initialization, In (pp. 130–140). Bei, Gray (b0060) 1985; 33 Gonzalez (b0145) 1985; 38 Mahajan, M., Nimbhorkar, P., & Varadarajan, K. (2012). The planar k-means problem is NP-hard. Bradley, P. S., & Fayyad, U. (1998). Refining initial points for k-means clustering, In > Garcia, Herrera (b0140) 2008; 9 He, J., Lan, M., Tan, C. L., Sung, S. Y., & Low, H. B. (2004). Initialization of cluster refinement algorithms: A review and comparative study, In Bergmann, Hommel (b0065) 1988 Hubert, Arabie (b0180) 1985; 2 Katsavounidis, Kuo, Zhang (b0215) 1994; 1 Su, Dy (b0345) 2007; 11 Astrahan, M. M. (1970). Speech analysis by clustering, or the hyperphoneme method, Tech. Rep. AIM-124, Stanford University. Redmond, Heneghan (b0325) 2007; 28 Wu, J., Xiong, H., & Chen, J. (2009b). Adapting the right measures for k-means clustering, In Kaufman, Rousseeuw (b0220) 1990 Milligan (b0285) 1980; 45 Hyvärinen (b0185) 1999; 10 SAS Institute Inc., SAS/STAT 9.2 User’s Guide, SAS Publishing, 2009. Luengo, Garcia, Herrera (b0245) 2009; 36 Garcia, Fernandez, Luengo, Herrera (b0135) 2009; 13 Matsumoto, Nishimura (b0275) 1998; 8 (pp. 487–494). Frank, A., & Asuncion, A. (2011). UCI machine learning repository, University of California, Irvine, School of Information and Computer Sciences Ordonez, Omiecinski (b0305) 2004; 16 Al Hasan, Chaoji, Salem, Zaki (b0015) 2009; 30 van Dongen, S. (2000). Performance criteria for graph clustering and Markov cluster experiments, Tech. Rep. INS-R0012, Centrum voor Wiskunde en Informatica. Jain, Murty, Flynn (b0200) 1999; 31 Hotelling (b0165) 1936; 1 Mao, Jain (b0270) 1996; 7 Wu, Chen, Xiong, Xie (b0370) 2009; 36 Cook, W. (2011). World TSP, Georgia Institute of Technology Dash, M., Liu, H., & Xu, X. (2001). ‘1 Bottou, Bengio (b0070) 1995 Hamerly, G. (2010). Making k-means even faster, in Onoda, Sakai, Yamada (b0300) 2012; 4 Breunig, Kriegel, Ng, Sander (b0080) 2000; 29 Forgy (b0120) 1965; 21 Friedman (b0130) 1937; 32 Jancey (b0205) 1966; 14 Al-Daoud, Roberts (b0010) 1996; 17 Pena, Lozano, Larranaga (b0315) 1999; 20 Kanungo, Mount, Netanyahu, Piatko, Silverman, Wu (b0210) 2002; 24 Selim, Ismail (b0335) 1984; 6 (pp. 91–99). Pal, Majumder (b0310) 1977; 7 Aloise, Hansen, Liberti (b0025) 2010 Ball, Hall (b0055) 1967; 12 Linde, Buzo, Gray (b0235) 1980; 28 Babu (10.1016/j.eswa.2012.07.021_b0050) 1994; 25 10.1016/j.eswa.2012.07.021_b0035 10.1016/j.eswa.2012.07.021_b0150 10.1016/j.eswa.2012.07.021_b0075 Matsumoto (10.1016/j.eswa.2012.07.021_b0275) 1998; 8 Jain (10.1016/j.eswa.2012.07.021_b0200) 1999; 31 Tarsitano (10.1016/j.eswa.2012.07.021_b0350) 2003; 36 Su (10.1016/j.eswa.2012.07.021_b0345) 2007; 11 Aloise (10.1016/j.eswa.2012.07.021_b0020) 2009; 75 10.1016/j.eswa.2012.07.021_b0115 Späth (10.1016/j.eswa.2012.07.021_b0340) 1977; 1 Milligan (10.1016/j.eswa.2012.07.021_b0290) 1988; 5 Mao (10.1016/j.eswa.2012.07.021_b0270) 1996; 7 Breunig (10.1016/j.eswa.2012.07.021_b0080) 2000; 29 Bei (10.1016/j.eswa.2012.07.021_b0060) 1985; 33 Hotelling (10.1016/j.eswa.2012.07.021_b0165) 1936; 1 10.1016/j.eswa.2012.07.021_b0160 Onoda (10.1016/j.eswa.2012.07.021_b0300) 2012; 4 Wu (10.1016/j.eswa.2012.07.021_b0370) 2009; 36 Bottou (10.1016/j.eswa.2012.07.021_b0070) 1995 Aloise (10.1016/j.eswa.2012.07.021_b0025) 2010 Lance (10.1016/j.eswa.2012.07.021_b0225) 1967; 10 Daniel (10.1016/j.eswa.2012.07.021_b0110) 2000 Gonzalez (10.1016/j.eswa.2012.07.021_b0145) 1985; 38 10.1016/j.eswa.2012.07.021_b0260 Jancey (10.1016/j.eswa.2012.07.021_b0205) 1966; 14 Tou (10.1016/j.eswa.2012.07.021_b0355) 1974 Zhang (10.1016/j.eswa.2012.07.021_b0380) 2003; 33 Celebi (10.1016/j.eswa.2012.07.021_b0090) 2011; 29 10.1016/j.eswa.2012.07.021_b0105 Huang (10.1016/j.eswa.2012.07.021_b0175) 1993; 2 Al Hasan (10.1016/j.eswa.2012.07.021_b0015) 2009; 30 Bergmann (10.1016/j.eswa.2012.07.021_b0065) 1988 Luengo (10.1016/j.eswa.2012.07.021_b0245) 2009; 36 Norušis (10.1016/j.eswa.2012.07.021_b0295) 2011 Al-Daoud (10.1016/j.eswa.2012.07.021_b0010) 1996; 17 Hubert (10.1016/j.eswa.2012.07.021_b0180) 1985; 2 10.1016/j.eswa.2012.07.021_b0255 10.1016/j.eswa.2012.07.021_b0330 10.1016/j.eswa.2012.07.021_b0375 Hartigan (10.1016/j.eswa.2012.07.021_b0155) 1979; 28 Hyvärinen (10.1016/j.eswa.2012.07.021_b0185) 1999; 10 Likas (10.1016/j.eswa.2012.07.021_b0230) 2003; 36 Selim (10.1016/j.eswa.2012.07.021_b0335) 1984; 6 Linde (10.1016/j.eswa.2012.07.021_b0235) 1980; 28 Pal (10.1016/j.eswa.2012.07.021_b0310) 1977; 7 Huang (10.1016/j.eswa.2012.07.021_b0170) 1990; 26 Pena (10.1016/j.eswa.2012.07.021_b0315) 1999; 20 Milligan (10.1016/j.eswa.2012.07.021_b0285) 1980; 45 Kanungo (10.1016/j.eswa.2012.07.021_b0210) 2002; 24 Lu (10.1016/j.eswa.2012.07.021_b0250) 2008; 29 Meilă (10.1016/j.eswa.2012.07.021_b0280) 2007; 98 Redmond (10.1016/j.eswa.2012.07.021_b0325) 2007; 28 10.1016/j.eswa.2012.07.021_b0365 10.1016/j.eswa.2012.07.021_b0320 Maitra (10.1016/j.eswa.2012.07.021_b0265) 2010; 19 Ordonez (10.1016/j.eswa.2012.07.021_b0305) 2004; 16 10.1016/j.eswa.2012.07.021_b0040 Forgy (10.1016/j.eswa.2012.07.021_b0120) 1965; 21 Jain (10.1016/j.eswa.2012.07.021_b0195) 2010; 31 10.1016/j.eswa.2012.07.021_b0360 Garcia (10.1016/j.eswa.2012.07.021_b0135) 2009; 13 Kaufman (10.1016/j.eswa.2012.07.021_b0220) 1990 10.1016/j.eswa.2012.07.021_b0005 10.1016/j.eswa.2012.07.021_b0125 Iman (10.1016/j.eswa.2012.07.021_b0190) 1980; 9 Garcia (10.1016/j.eswa.2012.07.021_b0140) 2008; 9 Chen (10.1016/j.eswa.2012.07.021_b0095) 2010; 18 Katsavounidis (10.1016/j.eswa.2012.07.021_b0215) 1994; 1 Cao (10.1016/j.eswa.2012.07.021_b0085) 2009; 58 Ball (10.1016/j.eswa.2012.07.021_b0055) 1967; 12 Babu (10.1016/j.eswa.2012.07.021_b0045) 1993; 14 Chen (10.1016/j.eswa.2012.07.021_b0100) 2011; 33 Friedman (10.1016/j.eswa.2012.07.021_b0130) 1937; 32 Anderberg (10.1016/j.eswa.2012.07.021_b0030) 1973 Lloyd (10.1016/j.eswa.2012.07.021_b0240) 1982; 28
References_xml	– volume: 19 start-page: 354 year: 2010 end-page: 376 ident: b0265 article-title: Simulating data to study performance of finite mixture modeling and clustering algorithms publication-title: Journal of Computational and Graphical Statistics – reference: Pizzuti, C., Talia, D., & Vonella, G. (1999). A divisive initialisation method for clustering algorithms, In: – reference: Cook, W. (2011). World TSP, Georgia Institute of Technology, – volume: 17 start-page: 451 year: 1996 end-page: 455 ident: b0010 article-title: New methods for the initialisation of clusters publication-title: Pattern Recognition Letters – volume: 29 start-page: 93 year: 2000 end-page: 104 ident: b0080 article-title: LOF: Identifying density-based local outliers publication-title: ACM SIGMOD Record – volume: 9 start-page: 2677 year: 2008 end-page: 2694 ident: b0140 article-title: An extension on “statistical comparisons of classifiers over multiple data sets” for all pairwise comparisons publication-title: Journal of Machine Learning Research – volume: 1 start-page: 27 year: 1936 end-page: 35 ident: b0165 article-title: Simplified calculation of principal components publication-title: Psychometrika – reference: (pp. 297–302). – volume: 2 start-page: 193 year: 1985 end-page: 218 ident: b0180 article-title: Comparing partitions publication-title: Journal of Classification – reference: (pp. 487–494). – volume: 38 start-page: 293 year: 1985 end-page: 306 ident: b0145 article-title: Clustering to minimize the maximum intercluster distance publication-title: Theoretical Computer Science – volume: 14 start-page: 127 year: 1966 end-page: 130 ident: b0205 article-title: Multidimensional group analysis publication-title: Australian Journal of Botany – volume: 14 start-page: 763 year: 1993 end-page: 769 ident: b0045 article-title: A near-optimal initial seed value selection in k-means algorithm using a genetic algorithm publication-title: Pattern Recognition Letters – reference: 1 – volume: 20 start-page: 1027 year: 1999 end-page: 1040 ident: b0315 article-title: An empirical comparison of four initialization methods for the k-means algorithm publication-title: Pattern Recognition Letters – volume: 30 start-page: 994 year: 2009 end-page: 1002 ident: b0015 article-title: Robust partitional clustering by outlier and density insensitive seeding publication-title: Pattern Recognition Letters – volume: 25 start-page: 85 year: 1994 end-page: 94 ident: b0050 article-title: Simulated annealing for selecting optimal initial seeds in the k-means algorithm publication-title: Indian Journal of Pure and Applied Mathematics – volume: 12 start-page: 153 year: 1967 end-page: 155 ident: b0055 article-title: A clustering technique for summarizing multivariate data publication-title: Behavioral Science – volume: 31 start-page: 264 year: 1999 end-page: 323 ident: b0200 article-title: Data clustering: A review publication-title: ACM Computing Surveys – volume: 8 start-page: 3 year: 1998 end-page: 30 ident: b0275 article-title: Mersenne twister: A 623-dimensionally equidistributed uniform pseudo-random number generator publication-title: ACM Transactions on Modeling and Computer Simulation – reference: > – volume: 18 start-page: 957 year: 2010 end-page: 966 ident: b0095 article-title: Bandwidth adaptive hardware architecture of k-means clustering for video analysis publication-title: IEEE Transactions on Very Large Scale Integration (VLSI) Systems – volume: 33 start-page: 983 year: 2003 end-page: 999 ident: b0380 article-title: Robust clustering by pruning outliers publication-title: IEEE Transactions on Systems, Man, and Cybernetics – Part B – volume: 32 start-page: 675 year: 1937 end-page: 701 ident: b0130 article-title: The use of ranks to avoid the assumption of normality implicit in the analysis of variance publication-title: Journal of the American Statistical Association – reference: Wu, J., Xiong, H., & Chen, J. (2009b). Adapting the right measures for k-means clustering, In: – volume: 33 start-page: 568 year: 2011 end-page: 586 ident: b0100 article-title: Parallel spectral clustering in distributed systems publication-title: IEEE Transactions on Pattern Analysis and Machine Intelligence – reference: (pp. 877–885). – volume: 28 start-page: 100 year: 1979 end-page: 108 ident: b0155 article-title: Algorithm AS 136: A k-means clustering algorithm publication-title: Journal of the Royal Statistical Society C – reference: Voss, T., Hansen, N., & Igel, C. (2010). Improved step size adaptation for the MO-CMA-ES, In: – reference: (pp. 1027–1035). – reference: (pp. 91–99). – reference: Hamerly, G. (2010). Making k-means even faster, in: – volume: 10 start-page: 271 year: 1967 end-page: 277 ident: b0225 article-title: A general theory of classificatory sorting strategies - II. Clustering systems publication-title: The Computer Journal – volume: 36 start-page: 451 year: 2003 end-page: 461 ident: b0230 article-title: The global k-means clustering algorithm publication-title: Pattern Recognition – reference: Bradley, P. S., & Fayyad, U. (1998). Refining initial points for k-means clustering, In: – reference: Frank, A., & Asuncion, A. (2011). UCI machine learning repository, University of California, Irvine, School of Information and Computer Sciences, – volume: 7 start-page: 625 year: 1977 end-page: 629 ident: b0310 article-title: Fuzzy sets and decision making approaches in vowel and speaker recognition publication-title: IEEE Transacations on Systems, Man, and Cybernetics – volume: 58 start-page: 474 year: 2009 end-page: 483 ident: b0085 article-title: An initialization method for the k-means algorithm using neighborhood model publication-title: Computers and Mathematics with Applications – volume: 7 start-page: 16 year: 1996 end-page: 29 ident: b0270 article-title: A self-organizing network for hyperellipsoidal clustering (HEC) publication-title: IEEE Transacations on Neural Networks – year: 1990 ident: b0220 article-title: Finding groups in data: An introduction to cluster analysis – volume: 9 start-page: 571 year: 1980 end-page: 595 ident: b0190 article-title: Approximations of the critical region of the friedman statistic publication-title: Communications in Statistics – Theory and Methods – volume: 45 start-page: 325 year: 1980 end-page: 342 ident: b0285 article-title: An examination of the effect of six types of error perturbation on fifteen clustering algorithms publication-title: Psychometrika – year: 1974 ident: b0355 article-title: Pattern recognition principles – volume: 29 start-page: 260 year: 2011 end-page: 271 ident: b0090 article-title: Improving the performance of k-means for color quantization publication-title: Image and Vision Computing – reference: He, J., Lan, M., Tan, C. L., Sung, S. Y., & Low, H. B. (2004). Initialization of cluster refinement algorithms: A review and comparative study, In: – volume: 98 start-page: 873 year: 2007 end-page: 895 ident: b0280 article-title: Comparing clusterings — An information based distance publication-title: Journal of Multivariate Analysis – volume: 36 start-page: 2955 year: 2003 end-page: 2966 ident: b0350 article-title: A computational study of several relocation methods for k-means algorithms publication-title: Pattern Recognition – reference: Dash, M., Liu, H., & Xu, X. (2001). ‘1 – reference: (pp. 130–140). – reference: (pp. 281–297). – volume: 2 start-page: 108 year: 1993 end-page: 112 ident: b0175 article-title: A comparison of several vector quantization codebook generation approaches publication-title: IEEE Transactions on Image Processing – reference: Mahajan, M., Nimbhorkar, P., & Varadarajan, K. (2012). The planar k-means problem is NP-hard. – reference: Arthur, D., & Vassilvitskii, S. (2007). k-means++: The advantages of careful seeding, In: – reference: 2’: Merging distance and density based clustering, In: – volume: 6 start-page: 81 year: 1984 end-page: 87 ident: b0335 article-title: K-means-type algorithms: A generalized convergence theorem and characterization of local optimality publication-title: IEEE Transactions on Pattern Analysis and Machine Intelligence – reference: (pp. 74–76). – reference: , – year: 2011 ident: b0295 article-title: IBM SPSS statistics 19 statistical procedures companion – reference: + – volume: 28 start-page: 84 year: 1980 end-page: 95 ident: b0235 article-title: An algorithm for vector quantizer design publication-title: IEEE Transactions on Communications – start-page: 1 year: 2010 end-page: 26 ident: b0025 article-title: An improved column generation algorithm for minimum sum-of-squares clustering publication-title: Mathematical Programming – volume: 1 start-page: 144 year: 1994 end-page: 146 ident: b0215 article-title: A new initialization technique for generalized Lloyd iteration publication-title: IEEE Signal Processing Letters – volume: 75 start-page: 245 year: 2009 end-page: 248 ident: b0020 article-title: NP-hardness of euclidean sum-of-squares clustering publication-title: Machine Learning – year: 1995 ident: b0070 article-title: Advances in neural information processing systems publication-title: Ch. convergence properties of the k-means algorithms – volume: 10 start-page: 626 year: 1999 end-page: 634 ident: b0185 article-title: Fast and robust fixed-point algorithms for independent component analysis publication-title: IEEE Transactions on Neural Networks – volume: 33 start-page: 1132 year: 1985 end-page: 1133 ident: b0060 article-title: An improvement of the minimum distortion encoding algorithm for vector quantization publication-title: IEEE Transactions on Communications – volume: 21 start-page: 768 year: 1965 ident: b0120 article-title: Cluster analysis of multivariate data: Efficiency vs. interpretability of classification publication-title: Biometrics – volume: 31 start-page: 651 year: 2010 end-page: 666 ident: b0195 article-title: Data clustering: 50 publication-title: Pattern Recognition Letters – volume: 28 start-page: 129 year: 1982 end-page: 136 ident: b0240 article-title: Least squares quantization in PCM publication-title: IEEE Transacations on Information Theory – reference: SAS Institute Inc., SAS/STAT 9.2 User’s Guide, SAS Publishing, 2009. – volume: 11 start-page: 319 year: 2007 end-page: 338 ident: b0345 article-title: In search of deterministic methods for initializing k-means and Gaussian mixture clustering publication-title: Intelligent Data Analysis – reference: van Dongen, S. (2000). Performance criteria for graph clustering and Markov cluster experiments, Tech. Rep. INS-R0012, Centrum voor Wiskunde en Informatica. – volume: 36 start-page: 6050 year: 2009 end-page: 6061 ident: b0370 article-title: External validation measures for k-means clustering: A data distribution perspective publication-title: Expert Systems with Applications – volume: 24 start-page: 881 year: 2002 end-page: 892 ident: b0210 article-title: An efficient k-means clustering algorithm: analysis and implementation publication-title: IEEE Transactions on Pattern Analysis and Machine Intelligence – volume: 4 start-page: 51 year: 2012 end-page: 59 ident: b0300 article-title: Careful seeding method based on independent components analysis for k-means clustering publication-title: Journal of Emerging Technologies in Web Intelligence – volume: 1 start-page: 23 year: 1977 end-page: 31 ident: b0340 article-title: Computational experiences with the exchange method: Applied to four commonly used partitioning cluster analysis criteria publication-title: European Journal of Operational Research – volume: 29 start-page: 787 year: 2008 end-page: 795 ident: b0250 article-title: Hierarchical initialization approach for k-means clustering publication-title: Pattern Recognition Letters – reference: (pp. 484–491). – reference: Astrahan, M. M. (1970). Speech analysis by clustering, or the hyperphoneme method, Tech. Rep. AIM-124, Stanford University. – reference: Al-Daoud, M. (2005). A new algorithm for cluster initialization, In: – volume: 36 start-page: 7798 year: 2009 end-page: 7808 ident: b0245 article-title: A study on the use of statistical tests for experimentation with neural networks: Analysis of parametric test conditions and non-parametric tests publication-title: Expert Systems with Applications – reference: MacQueen, J. (1967). Some methods for classification and analysis of multivariate observations, In: – volume: 16 start-page: 909 year: 2004 end-page: 921 ident: b0305 article-title: Efficient disk-based k-means clustering for relational databases publication-title: IEEE Transactions on Knowledge and Data Engineering – volume: 26 start-page: 1618 year: 1990 end-page: 1619 ident: b0170 article-title: Fast encoding algorithm for VQ-based image coding publication-title: Electronics Letters – reference: . – year: 1973 ident: b0030 article-title: Cluster analysis for applications – year: 1988 ident: b0065 article-title: Multiple hypotheses testing publication-title: Ch. improvements of general multiple test procedures for redundant systems of hypotheses – year: 2000 ident: b0110 article-title: Applied nonparametric statistics – volume: 13 start-page: 959 year: 2009 end-page: 977 ident: b0135 article-title: A study of statistical techniques and performance measures for genetics-based machine learning: Accuracy and interpretability publication-title: Soft Computing – volume: 5 start-page: 181 year: 1988 end-page: 204 ident: b0290 article-title: A study of standardization of variables in cluster analysis publication-title: Journal of Classification – reference: , 13-21. – reference: (pp. 32–39). – volume: 28 start-page: 965 year: 2007 end-page: 973 ident: b0325 article-title: A method for initialising the k-means clustering algorithm using kd-trees publication-title: Pattern Recognition Letters – volume: 7 start-page: 16 issue: 1 year: 1996 ident: 10.1016/j.eswa.2012.07.021_b0270 article-title: A self-organizing network for hyperellipsoidal clustering (HEC) publication-title: IEEE Transacations on Neural Networks doi: 10.1109/72.478389 – volume: 36 start-page: 7798 issue: 4 year: 2009 ident: 10.1016/j.eswa.2012.07.021_b0245 article-title: A study on the use of statistical tests for experimentation with neural networks: Analysis of parametric test conditions and non-parametric tests publication-title: Expert Systems with Applications doi: 10.1016/j.eswa.2008.11.041 – ident: 10.1016/j.eswa.2012.07.021_b0115 – volume: 14 start-page: 763 issue: 10 year: 1993 ident: 10.1016/j.eswa.2012.07.021_b0045 article-title: A near-optimal initial seed value selection in k-means algorithm using a genetic algorithm publication-title: Pattern Recognition Letters doi: 10.1016/0167-8655(93)90058-L – ident: 10.1016/j.eswa.2012.07.021_b0125 – volume: 31 start-page: 264 issue: 3 year: 1999 ident: 10.1016/j.eswa.2012.07.021_b0200 article-title: Data clustering: A review publication-title: ACM Computing Surveys doi: 10.1145/331499.331504 – volume: 36 start-page: 6050 issue: 3 year: 2009 ident: 10.1016/j.eswa.2012.07.021_b0370 article-title: External validation measures for k-means clustering: A data distribution perspective publication-title: Expert Systems with Applications doi: 10.1016/j.eswa.2008.06.093 – ident: 10.1016/j.eswa.2012.07.021_b0035 – volume: 12 start-page: 153 issue: 2 year: 1967 ident: 10.1016/j.eswa.2012.07.021_b0055 article-title: A clustering technique for summarizing multivariate data publication-title: Behavioral Science doi: 10.1002/bs.3830120210 – volume: 31 start-page: 651 issue: 8 year: 2010 ident: 10.1016/j.eswa.2012.07.021_b0195 article-title: Data clustering: 50years beyond k-means publication-title: Pattern Recognition Letters doi: 10.1016/j.patrec.2009.09.011 – volume: 16 start-page: 909 issue: 8 year: 2004 ident: 10.1016/j.eswa.2012.07.021_b0305 article-title: Efficient disk-based k-means clustering for relational databases publication-title: IEEE Transactions on Knowledge and Data Engineering doi: 10.1109/TKDE.2004.25 – volume: 33 start-page: 1132 issue: 10 year: 1985 ident: 10.1016/j.eswa.2012.07.021_b0060 article-title: An improvement of the minimum distortion encoding algorithm for vector quantization publication-title: IEEE Transactions on Communications doi: 10.1109/TCOM.1985.1096214 – volume: 36 start-page: 2955 issue: 12 year: 2003 ident: 10.1016/j.eswa.2012.07.021_b0350 article-title: A computational study of several relocation methods for k-means algorithms publication-title: Pattern Recognition doi: 10.1016/S0031-3203(03)00190-0 – volume: 28 start-page: 129 issue: 2 year: 1982 ident: 10.1016/j.eswa.2012.07.021_b0240 article-title: Least squares quantization in PCM publication-title: IEEE Transacations on Information Theory doi: 10.1109/TIT.1982.1056489 – volume: 19 start-page: 354 issue: 2 year: 2010 ident: 10.1016/j.eswa.2012.07.021_b0265 article-title: Simulating data to study performance of finite mixture modeling and clustering algorithms publication-title: Journal of Computational and Graphical Statistics doi: 10.1198/jcgs.2009.08054 – volume: 13 start-page: 959 issue: 10 year: 2009 ident: 10.1016/j.eswa.2012.07.021_b0135 article-title: A study of statistical techniques and performance measures for genetics-based machine learning: Accuracy and interpretability publication-title: Soft Computing doi: 10.1007/s00500-008-0392-y – volume: 2 start-page: 193 issue: 1 year: 1985 ident: 10.1016/j.eswa.2012.07.021_b0180 article-title: Comparing partitions publication-title: Journal of Classification doi: 10.1007/BF01908075 – volume: 75 start-page: 245 issue: 2 year: 2009 ident: 10.1016/j.eswa.2012.07.021_b0020 article-title: NP-hardness of euclidean sum-of-squares clustering publication-title: Machine Learning doi: 10.1007/s10994-009-5103-0 – ident: 10.1016/j.eswa.2012.07.021_b0160 – ident: 10.1016/j.eswa.2012.07.021_b0330 – volume: 29 start-page: 260 issue: 4 year: 2011 ident: 10.1016/j.eswa.2012.07.021_b0090 article-title: Improving the performance of k-means for color quantization publication-title: Image and Vision Computing doi: 10.1016/j.imavis.2010.10.002 – ident: 10.1016/j.eswa.2012.07.021_b0260 doi: 10.1016/j.tcs.2010.05.034 – volume: 28 start-page: 965 issue: 8 year: 2007 ident: 10.1016/j.eswa.2012.07.021_b0325 article-title: A method for initialising the k-means clustering algorithm using kd-trees publication-title: Pattern Recognition Letters doi: 10.1016/j.patrec.2007.01.001 – ident: 10.1016/j.eswa.2012.07.021_b0375 doi: 10.1145/1557019.1557115 – volume: 33 start-page: 983 issue: 6 year: 2003 ident: 10.1016/j.eswa.2012.07.021_b0380 article-title: Robust clustering by pruning outliers publication-title: IEEE Transactions on Systems, Man, and Cybernetics – Part B doi: 10.1109/TSMCB.2003.816993 – volume: 33 start-page: 568 issue: 3 year: 2011 ident: 10.1016/j.eswa.2012.07.021_b0100 article-title: Parallel spectral clustering in distributed systems publication-title: IEEE Transactions on Pattern Analysis and Machine Intelligence doi: 10.1109/TPAMI.2010.88 – ident: 10.1016/j.eswa.2012.07.021_b0105 – volume: 1 start-page: 23 issue: 1 year: 1977 ident: 10.1016/j.eswa.2012.07.021_b0340 article-title: Computational experiences with the exchange method: Applied to four commonly used partitioning cluster analysis criteria publication-title: European Journal of Operational Research doi: 10.1016/S0377-2217(77)81005-9 – volume: 30 start-page: 994 issue: 11 year: 2009 ident: 10.1016/j.eswa.2012.07.021_b0015 article-title: Robust partitional clustering by outlier and density insensitive seeding publication-title: Pattern Recognition Letters doi: 10.1016/j.patrec.2009.04.013 – volume: 25 start-page: 85 issue: 1–2 year: 1994 ident: 10.1016/j.eswa.2012.07.021_b0050 article-title: Simulated annealing for selecting optimal initial seeds in the k-means algorithm publication-title: Indian Journal of Pure and Applied Mathematics – volume: 11 start-page: 319 issue: 4 year: 2007 ident: 10.1016/j.eswa.2012.07.021_b0345 article-title: In search of deterministic methods for initializing k-means and Gaussian mixture clustering publication-title: Intelligent Data Analysis doi: 10.3233/IDA-2007-11402 – year: 1973 ident: 10.1016/j.eswa.2012.07.021_b0030 – volume: 9 start-page: 2677 year: 2008 ident: 10.1016/j.eswa.2012.07.021_b0140 article-title: An extension on “statistical comparisons of classifiers over multiple data sets” for all pairwise comparisons publication-title: Journal of Machine Learning Research – ident: 10.1016/j.eswa.2012.07.021_b0040 – volume: 17 start-page: 451 issue: 5 year: 1996 ident: 10.1016/j.eswa.2012.07.021_b0010 article-title: New methods for the initialisation of clusters publication-title: Pattern Recognition Letters doi: 10.1016/0167-8655(95)00119-0 – year: 2011 ident: 10.1016/j.eswa.2012.07.021_b0295 – start-page: 1 year: 2010 ident: 10.1016/j.eswa.2012.07.021_b0025 article-title: An improved column generation algorithm for minimum sum-of-squares clustering publication-title: Mathematical Programming – volume: 28 start-page: 100 issue: 1 year: 1979 ident: 10.1016/j.eswa.2012.07.021_b0155 article-title: Algorithm AS 136: A k-means clustering algorithm publication-title: Journal of the Royal Statistical Society C – ident: 10.1016/j.eswa.2012.07.021_b0320 doi: 10.1007/978-3-540-48247-5_62 – ident: 10.1016/j.eswa.2012.07.021_b0360 – volume: 10 start-page: 626 issue: 3 year: 1999 ident: 10.1016/j.eswa.2012.07.021_b0185 article-title: Fast and robust fixed-point algorithms for independent component analysis publication-title: IEEE Transactions on Neural Networks doi: 10.1109/72.761722 – volume: 7 start-page: 625 issue: 8 year: 1977 ident: 10.1016/j.eswa.2012.07.021_b0310 article-title: Fuzzy sets and decision making approaches in vowel and speaker recognition publication-title: IEEE Transacations on Systems, Man, and Cybernetics doi: 10.1109/TSMC.1977.4309789 – volume: 24 start-page: 881 issue: 7 year: 2002 ident: 10.1016/j.eswa.2012.07.021_b0210 article-title: An efficient k-means clustering algorithm: analysis and implementation publication-title: IEEE Transactions on Pattern Analysis and Machine Intelligence doi: 10.1109/TPAMI.2002.1017616 – ident: 10.1016/j.eswa.2012.07.021_b0255 – volume: 6 start-page: 81 issue: 1 year: 1984 ident: 10.1016/j.eswa.2012.07.021_b0335 article-title: K-means-type algorithms: A generalized convergence theorem and characterization of local optimality publication-title: IEEE Transactions on Pattern Analysis and Machine Intelligence doi: 10.1109/TPAMI.1984.4767478 – ident: 10.1016/j.eswa.2012.07.021_b0075 – volume: 18 start-page: 957 issue: 6 year: 2010 ident: 10.1016/j.eswa.2012.07.021_b0095 article-title: Bandwidth adaptive hardware architecture of k-means clustering for video analysis publication-title: IEEE Transactions on Very Large Scale Integration (VLSI) Systems doi: 10.1109/TVLSI.2009.2017543 – volume: 2 start-page: 108 issue: 1 year: 1993 ident: 10.1016/j.eswa.2012.07.021_b0175 article-title: A comparison of several vector quantization codebook generation approaches publication-title: IEEE Transactions on Image Processing doi: 10.1109/83.210871 – volume: 1 start-page: 27 issue: 1 year: 1936 ident: 10.1016/j.eswa.2012.07.021_b0165 article-title: Simplified calculation of principal components publication-title: Psychometrika doi: 10.1007/BF02287921 – volume: 14 start-page: 127 issue: 1 year: 1966 ident: 10.1016/j.eswa.2012.07.021_b0205 article-title: Multidimensional group analysis publication-title: Australian Journal of Botany doi: 10.1071/BT9660127 – volume: 36 start-page: 451 issue: 2 year: 2003 ident: 10.1016/j.eswa.2012.07.021_b0230 article-title: The global k-means clustering algorithm publication-title: Pattern Recognition doi: 10.1016/S0031-3203(02)00060-2 – volume: 21 start-page: 768 year: 1965 ident: 10.1016/j.eswa.2012.07.021_b0120 article-title: Cluster analysis of multivariate data: Efficiency vs. interpretability of classification publication-title: Biometrics – ident: 10.1016/j.eswa.2012.07.021_b0005 – volume: 58 start-page: 474 issue: 3 year: 2009 ident: 10.1016/j.eswa.2012.07.021_b0085 article-title: An initialization method for the k-means algorithm using neighborhood model publication-title: Computers and Mathematics with Applications doi: 10.1016/j.camwa.2009.04.017 – volume: 98 start-page: 873 issue: 5 year: 2007 ident: 10.1016/j.eswa.2012.07.021_b0280 article-title: Comparing clusterings — An information based distance publication-title: Journal of Multivariate Analysis doi: 10.1016/j.jmva.2006.11.013 – volume: 38 start-page: 293 issue: 2–3 year: 1985 ident: 10.1016/j.eswa.2012.07.021_b0145 article-title: Clustering to minimize the maximum intercluster distance publication-title: Theoretical Computer Science doi: 10.1016/0304-3975(85)90224-5 – volume: 20 start-page: 1027 issue: 10 year: 1999 ident: 10.1016/j.eswa.2012.07.021_b0315 article-title: An empirical comparison of four initialization methods for the k-means algorithm publication-title: Pattern Recognition Letters doi: 10.1016/S0167-8655(99)00069-0 – ident: 10.1016/j.eswa.2012.07.021_b0365 doi: 10.1145/1830483.1830573 – volume: 32 start-page: 675 issue: 200 year: 1937 ident: 10.1016/j.eswa.2012.07.021_b0130 article-title: The use of ranks to avoid the assumption of normality implicit in the analysis of variance publication-title: Journal of the American Statistical Association doi: 10.1080/01621459.1937.10503522 – year: 1974 ident: 10.1016/j.eswa.2012.07.021_b0355 – year: 1995 ident: 10.1016/j.eswa.2012.07.021_b0070 article-title: Advances in neural information processing systems – ident: 10.1016/j.eswa.2012.07.021_b0150 doi: 10.1137/1.9781611972801.12 – volume: 26 start-page: 1618 issue: 19 year: 1990 ident: 10.1016/j.eswa.2012.07.021_b0170 article-title: Fast encoding algorithm for VQ-based image coding publication-title: Electronics Letters doi: 10.1049/el:19901037 – volume: 1 start-page: 144 issue: 10 year: 1994 ident: 10.1016/j.eswa.2012.07.021_b0215 article-title: A new initialization technique for generalized Lloyd iteration publication-title: IEEE Signal Processing Letters doi: 10.1109/97.329844 – volume: 29 start-page: 93 issue: 2 year: 2000 ident: 10.1016/j.eswa.2012.07.021_b0080 article-title: LOF: Identifying density-based local outliers publication-title: ACM SIGMOD Record doi: 10.1145/335191.335388 – volume: 28 start-page: 84 issue: 1 year: 1980 ident: 10.1016/j.eswa.2012.07.021_b0235 article-title: An algorithm for vector quantizer design publication-title: IEEE Transactions on Communications doi: 10.1109/TCOM.1980.1094577 – volume: 5 start-page: 181 issue: 2 year: 1988 ident: 10.1016/j.eswa.2012.07.021_b0290 article-title: A study of standardization of variables in cluster analysis publication-title: Journal of Classification doi: 10.1007/BF01897163 – volume: 29 start-page: 787 issue: 6 year: 2008 ident: 10.1016/j.eswa.2012.07.021_b0250 article-title: Hierarchical initialization approach for k-means clustering publication-title: Pattern Recognition Letters doi: 10.1016/j.patrec.2007.12.009 – volume: 4 start-page: 51 issue: 1 year: 2012 ident: 10.1016/j.eswa.2012.07.021_b0300 article-title: Careful seeding method based on independent components analysis for k-means clustering publication-title: Journal of Emerging Technologies in Web Intelligence doi: 10.4304/jetwi.4.1.51-59 – year: 2000 ident: 10.1016/j.eswa.2012.07.021_b0110 – year: 1990 ident: 10.1016/j.eswa.2012.07.021_b0220 – volume: 9 start-page: 571 issue: 6 year: 1980 ident: 10.1016/j.eswa.2012.07.021_b0190 article-title: Approximations of the critical region of the friedman statistic publication-title: Communications in Statistics – Theory and Methods doi: 10.1080/03610928008827904 – year: 1988 ident: 10.1016/j.eswa.2012.07.021_b0065 article-title: Multiple hypotheses testing – volume: 10 start-page: 271 issue: 3 year: 1967 ident: 10.1016/j.eswa.2012.07.021_b0225 article-title: A general theory of classificatory sorting strategies - II. Clustering systems publication-title: The Computer Journal doi: 10.1093/comjnl/10.3.271 – volume: 8 start-page: 3 issue: 1 year: 1998 ident: 10.1016/j.eswa.2012.07.021_b0275 article-title: Mersenne twister: A 623-dimensionally equidistributed uniform pseudo-random number generator publication-title: ACM Transactions on Modeling and Computer Simulation doi: 10.1145/272991.272995 – volume: 45 start-page: 325 issue: 3 year: 1980 ident: 10.1016/j.eswa.2012.07.021_b0285 article-title: An examination of the effect of six types of error perturbation on fifteen clustering algorithms publication-title: Psychometrika doi: 10.1007/BF02293907
SSID	ssj0017007
Score	2.610252
Snippet	► K-means is the most widely used partitional clustering algorithm. ► k-means is highly sensitive to the selection of the initial centers. ► We present an... K-means is undoubtedly the most widely used partitional clustering algorithm. Unfortunately, due to its gradient descent nature, this algorithm is highly...
SourceID	proquest pascalfrancis crossref elsevier
SourceType	Aggregation Database Index Database Enrichment Source Publisher
StartPage	200
SubjectTerms	Algorithmics. Computability. Computer arithmetics Algorithms Applied sciences Cluster analysis Cluster center initialization Clustering Clusters Collection Computational efficiency Computer science; control theory; systems Data processing. List processing. Character string processing Exact sciences and technology Expert systems k-means Memory organisation. Data processing Partitional clustering Software Statistical tests Sum of squared error criterion Theoretical computing
Title	A comparative study of efficient initialization methods for the k-means clustering algorithm
URI	https://dx.doi.org/10.1016/j.eswa.2012.07.021 https://www.proquest.com/docview/1349469217 https://www.proquest.com/docview/1701023823
Volume	40
WOSCitedRecordID	wos000309378200019&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVESC databaseName: Elsevier SD Freedom Collection Journals 2021 customDbUrl: eissn: 1873-6793 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0017007 issn: 0957-4174 databaseCode: AIEXJ dateStart: 19950101 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3db9MwELdKxwMS4htRPiYj8ValcvNl5zFMgRVGhaYx9QEpch0bOrq0atqxf4T_l3Nsh2YTBR54iaprEqe9X3zn893vEHrFwAkOeMg9v4iJpz1obyqJ8hIVTwlPYjaVRd1sgo7HbDJJPnY6P1wtzMWcliW7vEyW_1XVIANl69LZf1B3c1MQwGdQOhxB7XD8K8WnNq_cMHpXjjRa1lwReud_pvOF-NwWYNoe0lWTb_jNO5dgv_pivtEkCnUR4_zLYjVbfz1vxfE1SfLaUkG7Irmt7fDGV_9wnPUPsqPs9agVfX0_Gr89Tk9r4SG48DDRNGg6zY5S49_qBgKzhf3GRidMWakJllnL3oo4UkCDacozkGa-BaR4MTVNEt2EbPibWsCzsyshW4baN_mw12yACUecDWT1XRNL6WAvHRBTh90m3L5iCJv0RJ_Cs7KE3UB7Po0S1kV76SibvGu2pygxdfjuF9lqLJM4eHXc33k8t5e8gvdQmQYq13yB2sE5uYfu2JUJTg2i7qOOLB-gu3aVgq0NqEDkGoE42UP0OcVbmMM15vBC4QZzuI05bDGHAXMYMIct5vAvzOEGc4_QpzfZycGhZ9t2eCKM_bV-4ZOhgkVbSJkICiFIQRLCSQHWoyCxXyQqkBFXImKKxJLFPAAJC8SQ-4KyIHiMuuWilE8QFiAhkYiUiMHSKDWNwyKCu0cBZ0L5tIeG7q_NheW0161V5rlLXjzLtTpyrY6c0BzU0UP95pqlYXTZeXbkNJZbn9T4mjnAbed1-y31NkM5bPXQS6fvHGZ0vU3HS7nYVLkmDA3jxB_SHedQzQWp9_Cf_mmgZ-iWX7dw0WHD56i7Xm3kC3RTXKxn1WrfIvsnB1rPRA
linkProvider	Elsevier
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=A+comparative+study+of+efficient+initialization+methods+for+the+k-means+clustering+algorithm&rft.jtitle=Expert+systems+with+applications&rft.au=EMRE+CELEBI%2C+M&rft.au=KINGRAVI%2C+Hassan+A&rft.au=VELA%2C+Patricio+A&rft.date=2013&rft.pub=Elsevier&rft.issn=0957-4174&rft.eissn=1873-6793&rft.volume=40&rft.issue=1&rft.spage=200&rft.epage=210&rft_id=info:doi/10.1016%2Fj.eswa.2012.07.021&rft.externalDBID=n%2Fa&rft.externalDocID=27095898
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0957-4174&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0957-4174&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0957-4174&client=summon