Improving the effectiveness of short text understanding by using web information mining

Short texts are always more difficult to understand. These short text are produced including social posts, conversations, keywords etc. and contains limited context. Short text consist more than one meaning and very complicated to understand. It consist limited amount of data because of that various methods like Text segmentation, part-of-speech tagging, and concept labeling are used. For better accuracy and result we used clustering. We use HMM(Hidden Markov Model)for text segmentation and POS tagging, Semantic matching algorithm for concept labeling as well as clustering algorithm (K-Means) for best results of Disambiguation. These all traditional methods are used for Short text understanding.