China
Africa
Explore chapters and articles related to this topic
Artificial Intelligence for Biomedical Informatics
Published in Ranjeet Kumar Rout, Saiyed Umer, Sabha Sheikh, Amrit Lal Sangal, Artificial Intelligence Technologies for Computational Biology, 2023
Shahid Azim, Samridhi Dev, Sushil Kumar, Aditi Sharan
Semantic matching techniques aim to compare two sentences to determine if they have a similar meaning. For example, the questions “Where do you sleep?” and “Where are you sleeping?” have almost the same words in them so we can say that they are asking the same question. But if you consider another question “where are you sleeping?” this also looks similar to the last question but has an entirely different meaning. Even in some cases, the words may totally not match but the questions are the same such as “How aged are you?” and “What is your age?” are exactly two same questions but have so common words. So in process of semantic matching, we train a network that returns a high similarity score when the questions are similar and a low similarity score when the questions are different.
Enhancing requirements reusability through semantic modeling and data mining techniques
Published in Enterprise Information Systems, 2018
Themistoklis Diamantopoulos, Andreas Symeonidis
The third limitation, which relates also to the possible inputs of our methodology, is that of the employed semantics. Although our approach involves structural and semantic matching (using ontologies and WordNet (Miller 1995)), it refrains from applying semantic inference on the ontologies, which could prove useful for validation purposes and possibly for revealing missing requirements information. Our approach instead focuses on functionality co-occurrence among requirements of different projects, therefore inference is performed at a data mining level, using association rule mining and model matching techniques. Though not suited for declarative methodologies that may require different level of semantics, our approach is however effective for imperative methodologies, while it can be used as a solid basis for further work. An indicative first effort towards this direction, which we have implemented for our semantic parsing module (Diamantopoulos et al. 2017), involves adding also inferred relationships (e.g. the phrase ‘the user can create his/her account’ includes not only an Action performed on ‘account’ but also Ownership of the ‘account’ by the ‘user’). Finally, a future extension in this aspect would also involve incorporating semantics in ontology level and/or further strengthening our semantic similarity methodology using methods such as salient semantic analysis (Luo et al. 2016) or by measuring similarity in higher orders of abstraction (Zhang et al. 2014).