Novelty detection is to retrieve new information and filter redundancy fromgiven sentences that are relevant to a specific topic. In TREC2003, the authors tried an approach tonovelty detection with semantic distance c...Novelty detection is to retrieve new information and filter redundancy fromgiven sentences that are relevant to a specific topic. In TREC2003, the authors tried an approach tonovelty detection with semantic distance computation. The motivation is to expand a sentence byintroducing semantic information. Computation on semantic distance between sentences incorporatesWordNet with statistical information. The novelty detection is treated as a binary classificationproblem: new sentence or not. The feature vector, used in the vector space model for classification,consists of various factors, including the semantic distance from the sentence to the topic and thedistance from the sentence to the previous relevant context occurring before it. New sentences arethen detected with Winnow and support vector machine classifiers, respectively. Several experimentsare conducted to survey the relationship between different factors and performance. It is provedthat semantic computation is promising in novelty detection. The ratio of new sentence size torelevant size is further studied given different relevant document sizes. It is found that the ratioreduced with a certain speed (about 0.86). Then another group of experiments is performedsupervised with the ratio. It is demonstrated that the ratio is helpful to improve the noveltydetection performance.展开更多
文摘Novelty detection is to retrieve new information and filter redundancy fromgiven sentences that are relevant to a specific topic. In TREC2003, the authors tried an approach tonovelty detection with semantic distance computation. The motivation is to expand a sentence byintroducing semantic information. Computation on semantic distance between sentences incorporatesWordNet with statistical information. The novelty detection is treated as a binary classificationproblem: new sentence or not. The feature vector, used in the vector space model for classification,consists of various factors, including the semantic distance from the sentence to the topic and thedistance from the sentence to the previous relevant context occurring before it. New sentences arethen detected with Winnow and support vector machine classifiers, respectively. Several experimentsare conducted to survey the relationship between different factors and performance. It is provedthat semantic computation is promising in novelty detection. The ratio of new sentence size torelevant size is further studied given different relevant document sizes. It is found that the ratioreduced with a certain speed (about 0.86). Then another group of experiments is performedsupervised with the ratio. It is demonstrated that the ratio is helpful to improve the noveltydetection performance.
