Skip links

Opinion Mining in Instagram Social Network with a case study of mobile phone product

Increasing the popularity and use of social networks such as Facebook, Instagram, and YouTube, which enables users to share their information and comments with others, requires a new tool for processing content generated by users. On most sales and promotion sites, there is a scoring section on the features of a product. But it is used to create a section of a questionnaire. The problem with the questionnaire is that, firstly, all aspects of designing a product may not be considered by the designer of the questionnaire, and secondly, people prefer to express their opinions in a few sentences, rather than filling out long-time questionnaires.

The purpose of this research is to provide a solution that, by using the opinion mining, firstly, without the need for a questionnaire, can create an automatic scoring system for mobile phone product features, and secondly, classifying different people’s comments into three groups: “positive opinion”, “negative opinion” and “neutral comments”.

The present study was conducted using 1000 English-language comments on the Instagram social network for the Samsung Galaxy S8. In the present study, the “Hu and Bing Liu” dictionary has been used and 10 features related to a mobile phone such as “camera” with their synonyms are defined. The results show that the combination of the use of the dictionary-based method and SVM machine learning algorithm is successful.

Join the Discussion

Return to top of page