A sentence classification technique using intention association expressions
The paper presents a technique of text classification based on the field association method. The authors extend this method to so-called intention association (IA) expressions in the communications and present a method of determining the speaker's intention for sentences in conversation. The formal rule descriptions using these IA expressions are defined in such a way that they can construct intention classification knowledge. The rules are weighted to enable identification of unnecessary sentences and classification of texts. As experimental material emails have been used in this research. The method uses a deterministic multi-attribute matching algorithm to find IA expressions from input sentences. The authors show in their experiments that this algorithm is 44.5 times faster than the well-known Aho-Corasick method. They also achieve very high precision and recall of unnecessary sentence extraction as well as of classification of each email.
- A document classification method by using field association words
- A new method for selecting English field association terms of compound words and its knowledge representation
- An improvement of the Aho-Corasick machine
- An information retrieval model based on vector space method by supervised learning
- Automatic summarization
- Efficient multi-attribute pattern matching
- Efficient string matching
- Extraction of field-coherent passages.
- scientific article; zbMATH DE number 3829300 (Why is no real title available?)
- Text categorization based on k-nearest neighbor approach for web site classification.
- The diversity-based approach to open-domain text summarization.
- The use of bigrams to enhance text categorization.
This page was built for publication: A sentence classification technique using intention association expressions
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5466759)
