Study of the Yahoo-Yahoo Hash-Tag Tweets Using Sentiment Analysis and Opinion Mining Algorithms

Mining opinion on social media microblogs presents opportunities to extract meaningful insight from the public trending issues like “yahoo-yahoo” which in Nigeria, is synonymous cybercrime. In this study, content analysis of selected historical tweets hash-tag was conducted for sentiment and topic modelling. A corpus 5500 obtained pre-processed using a pre-trained tweet tokenizer while Valence Aware Dictionary Sentiment Reasoning (VADER), Liu Hu method, Latent Dirichlet Allocation (LDA), Semantic Indexing (LSI) Multidimensional Scaling (MDS) graphs were used analysis, modelling visualization. Results showed had 173 unique clusters, 5327 duplicates frequency 9555 “yahoo”. Further validation mean scores ten volunteers returned R R2 0.8038 0.6402; 0.5994 0.3463; 0.5999 0.3586 Human VADER; Hu; VADER scores, respectively. While outperforms LDA LSI similar results The study confirms VADER’s performance unstructured data containing non-English slangs, conjunctions, emoticons, etc. proved that emojis are more representative sentiments than texts.