Semantic Similarity Computing Model Based on Multi Model Fine-Grained Nonlinear Fusion

Natural language processing (NLP) task has achieved excellent performance in many fields, including semantic understanding, automatic summarization, image recognition and so on. However, most of the neural network models for NLP extract text a fine-grained way, which is not conducive to grasp meaning from global perspective. To alleviate problem, combination traditional statistical method deep learning model as well novel based on multi nonlinear fusion are proposed this paper. The uses Jaccard coefficient part speech, Term Frequency-Inverse Document Frequency (TF-IDF) word2vec-CNN algorithm measure similarity sentences respectively. According calculation accuracy each model, normalized weight obtained results compared. weighted vector input into fully connected give final classification results. As result, sentence evaluation reduces granularity feature extraction, it can features globally. Experimental show that matching 84%, F1 value 75%.