Formulation of query statements by searchers for submission into relational databases and information retrieval systems have been a serious challenge that often lead to irrelevant search results. This is compounded by the level of uncertainty about the user's information need and in some cases, unfamiliarity with retrieval system. Evidently, the World Wide Web presents a more established challenge in this area, considering the fact that searchers has little or no training on search techniques on the web. This paper recognizes fuzzy logic system and fuzziness as a tool required to close the gap between automated systems and human thinking. We realize this stiffness in query presentation as against the flexibility in human thinking and then consider the fuzzy concept as a tool that can be incorporated into a new system to overcome the syntactic problem presented in most relational operations. Thus, the paper proposes a novel approach of natural language query based problems. We propose the use of an Unguided Loose Search (ULS) which involves the use of local appropriator on a fuzzified Natural Language Interface. Our approach incorporates fuzziness in the interface, using the local appropriator, of the database systems rather than within the data itself. It allows freedom to users since they will not have to learn any specialized syntax such as that of SQL. The result shows that the new querying model called the EFUSQL model is applicable to real life users and can be incorporated into existing databases and query interfaces. The results show that naïve users prefer the new system due to its flexibility and response time.
Online examination systems exist as an attempt to expand the frontiers of learning and testing. The system is aimed at taking the advantages of web resources to reduce time, cost and other constraints associated with location defined examination. The development has been limited by the strictness of the examination format vis-à-vis keyword match which are boundary-defined. This constraint led to poor adoption of online testing systems which are essential in handling the problem associated with large enrollment. Existing systems do not allow examiners to test for knowledge in ways that they seem due; similar to the conventional classroom testing system. We leveraged on the advances in Natural Language Processing and the success emerging from same and therefore remodeled the examination system against some known concepts in text summarization, term dependencies, semantic tagging and corporal build-up using a standalone global API for semantic interpretation of answers. We improve the Levenshtein distance between two strings a, b using the triangular inequality to identify the relationship of two terms as applicable between questions and answers. The Levenshtein distance was denoted in php with the int levenshtein ( string $str1 , string $str2 ). The output of the research is the development of a subjective examination system that allows the self-grading of essay type question using a web based semantic API. The complexity of the algorithm is O(m*n), where n and m are the length of str1 and str2 The interface was coded in php with a MySQL supporting backend.
Information Retrieval (IR) allows the identification of relevant information from connected repositories, however their performance have been of research interest leading to investigations in the modalities by which the accuracy of the retrievals are evaluated. Metrics such as Precision, Recall, F-score among others are used to evaluate an IR system. IR use same form of evaluation for both speech and text based system while failing to realize the difference that could have occurred in the process of transcription, especially in the voice to text search, which is the most common speech based search paradigm. This is forming a new set of concerns. This research aim to review and identify the strengths and weaknesses of existing metrics for measuring the performances of speech based. A total of 179 articles were retrieved using Google Scholar repository and were manually examined. Only 25 articles were selected for analysis in this study after applying our predefined inclusion and exclusion criteria. Result shows that Mean Average Precision is the most frequently used metric for speech based IR system with result range from 0.4191 to 0.620. Also transcription error of spoken query or spoken document has a near linear relationship to IR performance. This systematic review serves as a bibliography of speech based IR systems and can be used by those new to the field of IR.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.