GenCo: A Generative Learning Model for Heterogeneous Text Classification Based on Collaborative Partial Classifications

The use of generative learning models in natural language processing (NLP) has significantly contributed to the advancement applications, such as sentimental analysis, topic modeling, text classification, chatbots, and spam filtering. With a large amount generated each day from different sources, web-pages, blogs, emails, social media, articles, one most common tasks NLP is classification corpus. This important many institutions for planning, decision-making, creating archives their projects. Many algorithms exist automate but intriguing them that which also learns these automatically. In this study, we present new model infer learn data using probabilistic logic apply it classification. model, called GenCo, multi-input single-output (MISO) uses collaboration partial classifications generate desired output. It provides heterogeneity measure explain its results enables reduction curse dimensionality Experiments with were carried out on Twitter US Airline dataset, Conference Paper SMS Spam outperforming baseline 98.40%, 89.90%, 99.26% accuracy, respectively.