• REENA RAVEENDRAN
• Dr.D.MUHAMMAD NOORUL MUBARAK

The capacity to precisely decipher and evaluate text-based discussions is becoming more and more important in the quickly changing world of digital communication. The Chat Analysis System that is put forth here makes use of Global Vectors (GloVe) embeddings and Recurrent Neural Networks (RNNs) to comprehend the complex dynamics of these kinds of exchanges. This advanced method aims to accurately credit each communication to its correct sender by focusing on sender identification. Our technology, which is based on cutting-edge deep learning algorithms, can dissect, understand, and analyse text-based discussions in real-time, providing a reliable solution for a range of applications, from secure communications to social media surveillance. Previous research highlights the effectiveness of natural language processing (NLP) techniques in different domains. For instance, Muchhala et al. (2021) explored the use of bi-grams term frequency-inverse document frequency (TF-IDF) and probabilistic context-free grammar (PCFG) for fake news prediction. They demonstrated that these methods, combined with algorithms like Stochastic Gradient Descent, could identify unreliable sources with an accuracy of 77.2%. Similarly, Granik and Mesyura (2017) implemented a Naive Bayes classifier for fake news detection, achieving a classification accuracy of approximately 74% on a dataset of Facebook news posts. Their work underscores the potential of artificial intelligence methods in tackling complex textual analysis problems. Recent studies have further supported the use of advanced machine learning algorithms in text analysis. Shu et al.'s (2019) research highlighted the need for reliable models that can distinguish and categorise fake news while utilising social context and content to improve accuracy. Utilising stylometric traits for author attribution is another important contribution made by Rashkin et al. (2017). This is a crucial part of our proposed system's sender identification focus. All of this research supports the idea that advanced NLP and deep learning techniques can significantly increase the precision and dependability of text-based analysis Building on these foundational works, our Chat Analysis System aims to advance the field by integrating RNNs and GloVe embeddings to capture the semantic and syntactic nuances of conversations. This approach not only enhances the system's ability to understand the context but also improves the precision of sender identification. By addressing the gaps in existing methods and incorporating insights from previous research, our system aspires to set a new benchmark in the real-time analysis of text-based communications, ensuring both accuracy and efficiency.

GloVE,RNN,SVM,MLP,,RF,Precission,Recall,F1 Score

"GLOVE AND RNN CHAT ANALYZATION ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 6, page no.i14-i26, June-2024, Available :http://www.jetir.org/papers/JETIR2406802.pdf

"GLOVE AND RNN CHAT ANALYZATION ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 6, page no. ppi14-i26, June-2024, Available at : http://www.jetir.org/papers/JETIR2406802.pdf