• N.V.Ramya Devi Kotla
• V Jaya Lakshmi
• Y Nehasri
• G Navya

- One of the major components of social media is to connect people having similar opinions and interests. Also, the pioneering applications of social media that has an ability to dictate a person’s day to day life and a platform for people to share their opinion is Twitter. In twitter recommendations are based on hashtags, but it may not be the correct way of recommending tweets. The main objective of this work is to develop the recommendation system based on topic and sentiment. This will make group of people to share their opinions among them. For this work, we collected various datasets from Kaggle website and merged to form a new dataset. We implemented this work using LSTM model, Logistic Regression Model, SVM model. The accuracy obtained by LSTM model is 80%, which is more than Logistic Regression and SVM model. Among those methods LSTM model shows great accuracy when compared to other models. So, this system is successfully identifying the sentiment and topic of a tweet using LSTM model.

Twitter ,user personalization, sentiment analysis, topic modeling,LSTM

"Twitter User Personalization using NLP", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 4, page no.i41-i44, April-2024, Available :http://www.jetir.org/papers/JETIR2404806.pdf

"Twitter User Personalization using NLP", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 4, page no. ppi41-i44, April-2024, Available at : http://www.jetir.org/papers/JETIR2404806.pdf