Abstract:

The increasing spread of misinformation on online platforms necessitates effective detection mechanisms; thus, employing big data analytics with high-quality data is quite significant to enhance detection accuracy. Integration of these technologies enables intelligent and scalable processing of vast, diverse web content possible. Current approaches to disinformation detection can be plagued by a lack of contextual knowledge, low-quality data, and scalability issues, thus reducing their effectiveness. This paper proposes a framework integrating Natural Language Processing and Deep Learning (NLP-DL) particularly using Long Short-Term Memory (LSTM) networks, to process and interpret textual content with improved semantic comprehension, thus solving these problems. The proposed method enables real-time, large-scale disinformation detection through high-quality, labeled datasets and large data analysis. Results identify that the NLP-DL model reflects excellent accuracy and robustness against standard methods, thus projecting its potential application in systems of misinformation detection across numerous digital platforms