An Enhanced BERT Model for Depression Detection on Social Media Posts

Nareshkumar, R.; Nimala, K.

doi:10.1007/978-981-99-8479-4_5

Cited by 5 publications

(1 citation statement)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Data Pre-processing A series of important steps are required to prepare a Twitter dataset. These steps include data collection, text cleaning to remove noise such as URLs and special characters, tokenization for word separation, lowercase conversion for uniformity, stopword removal, normalization [19] using techniques such as stemming, emoticon handling, duplicate elimination, abbreviation expansion, sentiment analysis for emotional assessment, feature extraction, entity recognition and data labeling. These procedures, taken as a whole, transform the data into a format that is more organized and refined, making it more amenable to later analysis and applications of machine learning.…”

Section: Methodsmentioning

confidence: 99%

Graph-Based Rumor Detection on Social Media Using Posts and Reactions

et al. 2024

IJCDS

View full text Add to dashboard Cite

In this article, researchers deliver a novel method that makes use of graph-based contextual and semantic learning to detect rumors. Social media platforms are interconnected, so when an event occurs, similar news or user reactions with common interests are disseminated throughout the network. The presented research introduces an innovative graph-based method for identifying rumors on social media by analyzing both posts and reactions. Identifying and dealing with online rumors is an important and increasing difficulty. We use real-world social media data to create a solution based on data analysis. The process involves creating graphs, identifying bridge words, and selecting features. The proposed method shows better performance than the baselines, indicating its effectiveness in addressing this significant issue. The method that is being offered makes use of tweets and people's replies to them in order to comprehend the fundamental interaction patterns and make use of the textual and hidden information. The primary emphasis of this effort is developing a reliable graph-based analyzer that can identify rumors spread on social media. The modeling of textual data as a words co-occurrence graph results in the production of two prominent groups of significant words and bridge connection words. Using these words as building pieces, contextual patterns for rumor detection may be constructed and detected using node-level statistical measurements. The identification of unpleasant feelings and inquisitive components in the responses further enriches the contextual patterns. The recommended technique is assessed by means of the PHEME dataset, which is open to the public, and contrasted with a variety of baselines as well as our suggested approaches. The results of the experiments are encouraging, and the strategy that was suggested seems to be helpful for rumor identification on social media platforms online.

show abstract

Section: Methodsmentioning

confidence: 99%