발간년도 : [2022]
| 논문정보 |
|
| 논문명(한글) |
[Vol.17, No.1] Implementation of Visualization System Using Emotional Information of BERT-based Product Review |
|
| 논문투고자 |
Hyun-Ju Kim, So-Yeon Kim, Chang-Gun Kim |
|
| 논문내용 |
According to a recent survey by Embrain Trend Monitor, 86.9% of 1,200 men and women between the ages of 19-49 needed a product review. Among them, 78.6% of all respondents said that they always check consumer reviews when purchasing products. As such, in recent years, as the importance of product reviews increases, various attempts are being introduced to improve reviews. Representative domestic examples include Daily Hotel's True Review, CGV's Golden Egg Index, Play Store's popular function, and Mango Plate's Dining Code. The positive and negative intentions of consumers implied in these reviews can become factors that influence other consumers' purchasing decisions in the future. Therefore, in this paper, a system for visualizing emotional information for product reviews was proposed using the BERT model applied with deep learning techniques. The proposed system consists of three modules: a web crawling module that collects basic information about food products from the web, a module that creates an emotional information model of reviews collected by using the BERT model, and a module that visualizes product information based on the created emotional information. It is composed of subsystems. The system proposed in this paper obtained an excellent measurement value of 93.8 in accuracy measurement, which is one of the performance evaluation indicators of the BERT model. It is expected that this will be applied to various fields that utilize positive/negative evaluation of the food product review. |
|
| 첨부논문 |
|
|
|
|
|
자료실 > 논문지 
