Loading...

Extraction and Analysis of Product Aspects of Online Stores Based on Customers' Reviews Using Machine Learning Techniques

Kazemi Foroushani, Amir Hossein | 2023

61 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 56474 (01)
  4. University: Sharif University of Technology
  5. Department: Industrial Engineering
  6. Advisor(s): Habibi, Moslem; Fazli, Mohammad Amin
  7. Abstract:
  8. Today, with the popularity of online shopping, many business owners are interested in launching online stores. Recording opinions and using the opinions of customers and buyers of products provides the possibility of listening to the opinions of customers in the design and improvement of products. On the other hand, other users and those who intend to buy a specific product, after paying attention to the features and aspects of the product they want, refer to the section of customer views and opinions. The user also reads these comments among thousands of different sentences and words written with different opinions and suggests; It gets confusing and many users won't be able to make the right decision after spending a lot of time in the comments section. In summary, it can be said that the purpose of this research is to fulfill a need, the answer of which is in the analysis of data and text views of online store users using the combination of techniques and methods of Machine learning and sentiment analysis. On the other hand, by extracting the features of the products, it will be possible to search and sort the identified features of each product. This research uses the rules and structure of Persian language and relationships between words to identify the aspects or features mentioned in the text. Therefore, the presented algorithm can be extended to other opinions and viewpoints submitted for any product, service, organization and other entities. The presented algorithm covers the three stages of extracting views, identifying and extracting aspects, and finally analyzing the sentiments of each aspect. In order to evaluate the results we use the largest Persian-language online store
  9. Keywords:
  10. Sentiment Analysis ; Aspect Mining ; Opinion Mining ; Aspect Based Sentiment Analysis ; Explicit Aspect ; Implicit Aspect ; Machine Learning ; Online Store

 Digital Object List

 Bookmark

...see more