RESEARCH ON OPTIMAL SELECTION STRATEGY OF SEARCH ENGINE KEYWORDS BASED ON 0-1 KNAPSACK

Keyword advertising business is an extremely important source of profit for search engine companies. Search engine companies have designed a complete set of keyword bidding schemes for search advertising, in which advertisers participate in keyword bidding. Advertisers need to select a certain numbe...

Full description

Saved in:
Bibliographic Details
Published in:Scientific Bulletin. Series C, Electrical Engineering and Computer Science no. 4; p. 79
Main Author: Nie, Han
Format: Journal Article
Language:English
Published: Bucharest University Polytechnica of Bucharest 01.01.2023
Subjects:
Advertising
Budgets
Dynamic programming
Keywords
Optimization models
Revenue
Search engines
ISSN:2286-3540
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Keyword advertising business is an extremely important source of profit for search engine companies. Search engine companies have designed a complete set of keyword bidding schemes for search advertising, in which advertisers participate in keyword bidding. Advertisers need to select a certain number of keywords, hoping to gain maximum profit. With limited advertising budget resources, it is impossible for advertisers to evenly allocate resources to each keyword. How to reasonably select keywords from the existing set of candidate keywords to maximize overall revenue is a problem faced by every advertiser. This paper will set up a keyword optimization model based on 0-1 knapsack. In this paper dynamic programming is used to attain the result. The model proposed in this article helps advertisers select keywords through their performance parameters (such as search demand, click rate, cost per click, etc.). This paper designs an experimental scenario. Firstly, an initial set of keywords is obtained through keyword mining tools, and then the model in this paper is used to solve and provide keyword selection results. The analysis of experimental results shows that the model proposed in this paper can recommend a keyword set with a total revenue superior to other methods within the budget range.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2286-3540