Multimodal rule transfer into automatic knowledge based topic models

Topic models are used in text analysis to extract domain features and to explore unknown domains. The topic models and its extensions follow traditional machine learning approach as single-shot learning. Automatic knowledge based topic models (AKBTM) filled this gap by learning from each task and carrying it to future tasks as knowledge rules. Most of the research in AKBTM focuses on rule extraction techniques. The transfer of rules is ignored for the most part of it, using single transfer mode. In this research paper, a multimodal rule transfer mechanism is proposed that operate in three modes to transfer the impact of rules into the inference technique. The rules are divided into two bins, based on their quality, as correlation strength. The mode of transferring rules bias is governed by the strength of rule and time phase the inference technique is in. It resulted in 24 points of improvement in topic coherence of the proposed model as compared to state of the art. The efficient and appropriate transfer of rule bias into the model, helped improve performance by 38%.