Leveraging Web Semantic Knowledge in Word Representation Learning

Abstract

Much recent work focuses on leveraging semantic lexicons like WordNet to enhance word representation learning (WRL) and achieves promising performance on many NLP tasks. However, most existing methods might have limitations because they require high-quality, manually created, semantic lexicons or linguistic structures. In this paper, we propose to leverage semantic knowledge automatically mined from web structured data to enhance WRL. We first construct a semantic similarity graph, which is referred as semantic knowledge, based on a large collection of semantic lists extracted from the web using several pre-defined HTML tag patterns. Then we introduce an efficient joint word representation learning model to capture semantics from both semantic knowledge and text corpora. Compared with recent work on improving WRL with semantic resources, our approach is more general, and can be easily scaled with no additional effort. Extensive experimental results show that our approach outperforms the state-of-the-art methods on word similarity, word sense disambiguation, text classification and textual similarity tasks.

Clustering Fake News with K-Means and Agglomerative Clustering Based on Word2Vec

“…By leveraging the semantic information embedded in Word2Vec representations, The goal of the suggested method is to improve clustering's accuracy and effectiveness, contributing to the development of robust tools for detecting and organizing Fake News content. The major reason why distributed word representations improve performance in numerous NLP applications is their ability to Izhar Muhammad Tianda 1 , IJMCR Volume 12 Issue 02 February 2024 capture semantic regularities [2]. The outcomes of this research are anticipated to offer insights into the potential application of clustering algorithms for effective Fake News identification, thereby contributing to the ongoing efforts to mitigate the adverse effects of misinformation in the digital age.…”