| Title |
A Method for Predicting the Number of Visitors of Keyword Search Advertising based on Bid Change Data using ANN |
| Authors |
이정우(Jung-Woo Lee) ; 김승천(Seung-Cheon Kim) |
| DOI |
https://doi.org/10.5573/ieie.2021.58.12.42 |
| Keywords |
Online advertising; Search advertising; Advertising performance prediction; RTB; ANN |
| Abstract |
Online advertising is very important for small and medium sized e-commerce shopping malls. According to statistics, the preference and effectiveness of keyword search advertising are high for both advertisers and customers. Small and medium sized e-commerce shopping malls are advertisers with a small advertising budget, and for them, an advertising performance prediction model is needed for advertising operation that can achieve high advertising performance with low advertising costs by increasing advertising efficiency. The search advertising performance prediction model has been studied extensively with various artificial intelligence modeling techniques. However, most studies predict performance using user information data and purchase history. In an environment where personal information protection is strengthened worldwide, it is difficult to predict performance using customer personal information data, so research on advertising performance prediction using new types of data is needed. In this study, using the search keyword bid change log data stored through the operation of online search advertising and the number of visitors logs data as the advertising performance data, a program to predict the number of visitors among search advertisement performance was implemented using ANN among artificial intelligence modeling techniques. The performance was measured using actual advertising operation data of small and medium sized e-commerce shopping malls. |