The Domain Restriction Hypothesis: Relating Term Similarity and Semantic Consistency
نویسندگان
چکیده
In this paper, we empirically demonstrate what we call the domain restriction hypothesis, claiming that semantically related terms extracted from a corpus tend to be semantically coherent. We apply this hypothesis to define a post-processing module for the output of Espresso, a state of the art relation extraction system, showing that irrelevant and erroneous relations can be filtered out by our module, increasing the precision of the final output. Results are confirmed by both quantitative and qualitative analyses, showing that very high precision can be reached.
منابع مشابه
A Relative Structure Similarity Method For Multiple Ontologies Alignment
Knowledge in domain is expressed with the help of ontology which is scattered all over its space. Using ontology gives a share in increasing precision. Different ontologies may represent the same domain, thus includes different terms that equivalently refer to the same meaning and vice versa. This results in different structures for ontologies. That's why it is necessary to relate concepts and ...
متن کاملA Comparative Study of Ontology Based Term Similarity Measures on PubMed Document Clustering
Recent research shows that ontology as background knowledge can improve document clustering quality with its concept hierarchy knowledge. Previous studies take term semantic similarity as an important measure to incorporate domain knowledge into clustering process such as clustering initialization and term re-weighting. However, not many studies have been focused on how different types of term ...
متن کاملConfidence Measure Based on Context Consistency Using Word Occurrence Probability and Topic Adaptation for Spoken Term Detection
In this paper, we propose a novel confidence measure to improve the performance of spoken term detection (STD). The proposed confidence measure is based on the context consistency between a hypothesized word and its context in a word lattice. The main contribution of this paper is to compute the context consistency by considering the uncertainty in the results of speech recognition and the effe...
متن کاملSemantic-Based Image Retrial in the VQ Compressed Domain using Image Annotation Statistical Models
متن کامل
A Comparative and Contrastive Study on the Meaning Extension of Color Terms in Persian and English
We deal with a wide range of colors in our daily life. They are such ubiquitous phenomena that is hard and next to impossible to imagine even a single entity (be it an object, place, living creature, etc) devoid of them. They are like death and tax which nobody can dispense with. This omnipresence of colors around us has also made its way through abstract and less tangible entities via the inte...
متن کامل