The study focuses on applying various clustering algorithms to analyze the Lok Sabha election results from the Election Commission of India. The primary objective is to identify patterns and insights that can help understand voter behavior, demographic influences, and regional political dynamics. The dataset comprises detailed election results, including vote counts, winning margins, and party affiliations, across multiple election cycles.
We implemented several clustering algorithms, including K-means, DBSCAN, and hierarchical clustering, to group constituencies based on voting patterns and demographic factors. The performance of these algorithms was evaluated using metrics such as silhouette scores, Davies-Bouldin index, and cluster coherence. The analysis revealed distinct clusters of constituencies with similar voting behaviors and demographic profiles, providing a nuanced understanding of regional political trends. With its simplicity and efficiency, K-means clustering provided clear and interpretable clusters, whereas DBSCAN effectively identified outliers and dense data regions. Hierarchical clustering offered insights into the nested structure of constituency similarities. Our findings highlight the importance of demographic factors such as literacy rate, urbanization, and socio-economic status in shaping electoral outcomes. The clustering results can aid political analysts, campaign strategists, and policymakers in formulating targeted strategies for future elections.
In conclusion, the application of clustering algorithms to Lok Sabha election results demonstrates the potential of machine learning techniques in political data analysis. Future work could involve the integration of additional data sources, such as social media sentiments and economic indicators, to enhance the robustness of the clustering models.
my presentation was okok. well not really. but my professor couldnt believe all chunks of code. for some reason he failed to understand how a jupyter notebook works. he ended up turning it down in the start lmao but then understood it well