• ISSN: 2349-6002
  • UGC Approved Journal No 47859

Personalized Mobile App Recommendation by Learning User’s Interest from Social Media

  • Unique Paper ID: 155162
  • Volume: 9
  • Issue: 1
  • PageNo: 65-70
  • Abstract:
  • The knowledge and interests of users are valuable resources in social media, but they are frequently neglected. The vast majority of users on social media are not tagged, which means that their interests and areas of expertise are effectively hidden from useful applications such as personalised recommendation, community detection, and expert mining. For example, social media platforms provide only a partial view of what users are known for (by aggregating the knowledge of crowd tagging), but this information is still available. By learning the interest’s association between applications and posts, we are able to generate personalised app recommendations for our users. To do this, we employ a one-of-a-kind generative model known as IMCF+, which converts user interest from rich post information to sparse app usage. We conducted an analysis of the performance of this method, which predicts the top ten apps with an accuracy of 82.5% despite requiring only 10% of the data for training purposes. In addition, this purpose technique outperforms the other six state-of-the-art algorithms by 4.7 percent and 10 percent respectively in the high sparsity situation and the user cold- start scenario, which demonstrates the efficacy of our technology. All of these findings indicate that our method is capable of reliably extracting user interests from posts in order to contribute to the solution of the problem of providing personalised app recommendations.

Cite This Article

  • ISSN: 2349-6002
  • Volume: 9
  • Issue: 1
  • PageNo: 65-70

Personalized Mobile App Recommendation by Learning User’s Interest from Social Media

Related Articles