A Comparative Analysis of Social Communication Applications using Aspect Based Sentiment Analysis
| Autor: | Akmal Khan, Yang Yu, Muhammad Ayoub, Shabir Hussain, Laiba Irfan |
|---|---|
| Rok vydání: | 2022 |
| Zdroj: | Pakistan Journal of Engineering and Technology. 5:44-50 |
| ISSN: | 2664-2050 2664-2042 |
| DOI: | 10.51846/vol5iss3pp44-50 |
| Popis: | Google Play Store is a popular distribution channel with millions of applications. WhatsApp is the most downloaded communication application on Play Store. A few months ago, WhatsApp changed its privacy policy, triggering a wave of user reviews outrage. Privacy is essential in the application; users are worried about their data security and privacy. A computational system must be required to analyze the user’s reviews for WhatsApp authority to make better policies. This study aims to develop a deep learning-based model for automatically assessing reviews that can be adapted for future data analysis. We proposed a deep learning methodology by using Aspect-based sentiment analysis (ABSA) utilizing the communication app reviews scraped from the Google play store using the Google Play scrapper application. This study uses the text mining technique for ABSA on the user’s reviews. For Topic extraction, we have used Latent Dirichlet Allocation (LDA) and the deep learning method Long Short-Term Memory (LSTM) for topic classification. The results show that our proposed model gives us a promising outcome with 90% accuracy by using the LSTM model. WhatsApp authority can use the results to optimize communication applications by adding more efficient features and updating them. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|