Stance Detection using Extreme Learning Machine with Improved Agglomerative Hierarchical Clustering on Social Network
Saini Jacob Soman1, R. Anandan2
1Saini Jacob Soman, Research Scholar, Department of Computer Science and Engineering, Vels Institute of Science, Technology & Advanced Studies, Pallavaram, Chennai, India.
2R. Anandan, Professor, Department of Computer Science and Engineering, Vels Institute of Science, Technology & Advanced Studies, Pallavaram, Chennai, India.

Manuscript received on November 11, 2019. | Revised Manuscript received on November 20 2019. | Manuscript published on 30 November, 2019. | PP: 11244-11252 | Volume-8 Issue-4, November 2019. | Retrieval Number: D9426118419/2019©BEIESP | DOI: 10.35940/ijrte.D9426.118419

Open Access | Ethics and Policies | Cite  | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: People can communicate and exchange their views through online social networks and it laid the host for the online social groups. Among numerous groups: privacy violation, groups with no choices of opt-in, disorder were the main problems, which inhibit security of the user and comfort, and we consider user as the member so managing the group principles becomes tedious one. The stylistic, thematic, emotional, sentimental paves the ways for clustering the posts within the group and psycholinguistic rectify these major problems. Stance detection has recently gained significant attention in research and it is sentimental clustering, further it forms a primary segment of the larger research challenge posed from Facebook groups. Recognizing the position of certain Facebook, with regard to the topics given, from user-made text was the main issue addressed in this work. Preprocessing, keyword extraction, stance detection using ELM and clustering using IAHC algorithm were there in this system. Initial one (pre-processing) helps to eliminate the unnecessary data and further assist us to enhance the clustering accuracy in the given dataset. Then for choosing the prominent keywords based on the frequent terms, the keyword extraction was done. Then for classifying neutral and non-neutral posts, Extreme Learning Machine (ELM) paves the way. Further, personal posts are categorized to be positive vs. negative. The members of the group were classified, based on the response to the post of various features. This classification fulfills the performance of huge members present in a strong group by executing the clustering on the basis of linguistic characteristics. For providing sentiments that depend on the posts and the response given by users for the posts, then enhancing the performance of the system, Improved Agglomerative Hierarchical Clustering (IAHC) was established. As observed from the experimental analysis, it is proven that the proposed ELM with IAHC algorithm yields higher performance when compared with the existing methods.
Keywords: Stance Detection, Extreme Learning Machine, Social Network, Sentiment Analysis, Improved Agglomerative Hierarchical Clustering.
Scope of the Article: Machine Learning.