International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 04 | Apr 2025
p-ISSN: 2395-0072
www.irjet.net
Instagram Comment Classification by Using Logistic Regression Based SSK-Mean Clustering Technique P JAGAN MOHAN1, S MUNI KUMAR2 1Student, Dept of MCA, KMMIPS, Tirupati
2Associate Professor, Dept of MCA, KMMIPS, Tirupati
---------------------------------------------------------------------***--------------------------------------------------------------------1.1 Clustering uses for machine learning. Abstract - Text classification is a fundamental task in natural language processing (NLP) that aims to categorize textual data into predefined classes. In this paper, we propose a hybrid approach that combines Logistic Regression (LR) with the SSK-Means Clustering Algorithm to improve the efficiency and accuracy of text classification. Logistic Regression, a widely used supervised learning technique, provides a robust probabilistic framework for text classification. However, to enhance the model’s performance, we integrate it with the SSK-Means (String Subsequence Kernel K-Means) Clustering Algorithm, which leverages string subsequence kernels to capture the semantic similarity between text samples. The SSK-Means algorithm enables better feature representation by grouping similar textual data before classification, thereby reducing noise and improving classification accuracy. Experimental results on benchmark datasets demonstrate that our proposed method outperforms traditional approaches in terms of precision, recall, and F1-score. This hybrid model offers an effective solution for text classification tasks in various real-world applications, including sentiment analysis, spam detection, and topic categorization.
(1) Machine learning is important in adjusting its structure to produce desired outputs due to the heavy amount of data input into the system (2) Machine learning is also suitable for data mining because of the little amount of important data hidden in the heavy chunk of data that can be important for processing of output. (3) Machine learning is important for jobs that are on the go thereby improving the existing machine designs because some designers produce non-workable machines that are not desired in the environment.[2] Harwath, D., Torralba, A., Glass, J. Unsupervised l ea rnin g of spok en l a ngua ge with visual context. In Advances in Neural Information Processing Systems, 2016: 1858-1866.
1.2 Clustering Clustering is an unsupervised learning technique used to categorize patterns (observations, data points, or feature vectors) into distinct groups (clusters) based on similarity. It is widely applied in various domains as a fundamental step in exploratory data analysis. Despite its broad applicability, clustering remains a challenging combinatorial problem. Variations in assumptions, methodologies, and application contexts across different disciplines have slowed the transfer of universal clustering concepts and techniques.
Key Words: machine learning, clustering, logistic regression, unsupervised learning, supervised learning, data analysis, prediction, classification, segmentation, data mining, model evaluation, algorithms, and applications
1.INTRODUCTION
A. K. Jain, M. N. Murty, and P. J. Flynn. 1999. Data clustering: a review. ACM Compute. Surv. 31, 3 (Sept. 1999), 264–323.
Logistic Regression is a widely used supervised learning algorithm for classification tasks. When applied to text-based classification, it helps categorize textual data into predefined classes .Simple and efficient for binary and multi-class text classification. Works well with sparse data (which is common in text representation).Uses probabilities (Sigmoid function) to determine class membership. Text data is converted into a numerical format using TF-IDF Vectorization or Word Embeddings. The transformed text data is fed into a Logistic Regression model.The model learns to classify text based on features extracted from words, phrases, or entire sentences.
© 2025, IRJET
|
Impact Factor value: 8.315
Clustering is a widely used analytical technique for grouping unlabeled data to extract meaningful insights. Since no single clustering algorithm can address all clustering problems, various algorithms have been developed for diverse applications. It is defined as the process of grouping objects when there is little or no prior knowledge about their relationships in the given dataset. Clustering also aims to uncover the underlying patterns or classes present within the data. Additionally, it serves as a method for organizing unlabeled data into distinct groups with minimal or no supervision.
|
ISO 9001:2008 Certified Journal
|
Page 488