Research on the Calculation Method of Semantic Similarity Based on Concept Hierarchy

Hybridization model of frame semantics and deep learning for text semantic similarity calculation

“…The dependency information is combined with the semantic information vector and the location information vector to form a multi-feature combination data. Similarly, literature [4] proposed the application of structured features to represent syntactic and semantic information of sentence-level text to address the problem of weak representation of flat features in sentence-level text similarity calculation methods. The literature [6] integrates various textual features such as semantics, syntax and word frequency, and constructs a textual complex network integrating co-occurrence distance and dependent syntax, and uses information entropy to determine the weights of network dynamics indicators.…”

“…In recommendation systems, by analysing users' historical behaviours and preferences, recommendation systems can use text semantic similarity algorithms to recommend content that better matches users' interests [3] . In the field of public opinion monitoring, by analysing a large amount of text data such as social media and news reports, text semantic similarity algorithms can help enterprises, governments and other organizations understand public attitudes and emotions towards a certain event or topic [4] .…”