on unseen problem instances as well as other problem domains desirably with no additional human expert intervention. Hence, the proposed method is additionally applied to a high school timetabling problem, as well as six other problem domains from a hyper-heuristic benchmark to test its level of generality. The empirical results show that our easy-to-implement hyper-heuristic is effective in solving the Yeditepe course timetabling problem. Moreover, being sufficiently general, it delivers a reasonable performance across different problem domains.
The course timetabling problem is a well known constraint optimization problem which has been of interest to researchers as well as practitioners. Due to the NP-hard nature of the problem, the traditional exact approaches might fail to find a solution even for a given instance. Hyper-heuristics which search the space of heuristics for high quality solutions are alternative methods that have been increasingly used in solving such problems. In this study, a curriculum based course timetabling problem at Yeditepe University is described. An improvement oriented heuristic selection strategy combined with a simulated annealing move acceptance as a hyper-heuristic utilizing a set of low level constraint oriented neighbourhood heuristics is investigated for solving this problem. The proposed hyper-heuristic was initially developed to handle a variety of problems in a particular domain with different properties considering the nature of the low level heuristics. On the other hand, a goal of hyper-heuristic development is to build methods which are general. Hence, the proposed hyper-heuristic is applied to six other problem domains and its performance is compared to different state-of-the-art hyper-heuristics to test its level of generality. The empirical results show that the proposed method is sufficiently general and powerful
The emergence of an ever increasing number of documents makes it more and more difficult to locate them when desired. An approach for improving search results is to make use of user-generated tags. This approach has led to improvements. However, they are limited because tags are (1) free from context and form, (2) user generated, (3) used for purposes other than description, and (4) often ambiguous. As a formal, declarative knowledge representation model, Ontologies provide a foundation upon which machine understandable knowledge can be obtained and tagged, and as a result, it makes semantic tagging and search possible. With an ontology, semantic web technologies can be utilized to automatically generate semantic tags. WordNet has been used for this purpose. However, this approach falls short in tagging documents that refer to new concepts and instances. To address this challenge, we present UNIpedia -a platform for unifying different ontological knowledge bases by reconciling their instances as WordNet concepts. Our mapping algorithms use rule based heuristics extracted from ontological and statistical features of concept and instances. UNIpedia is used to semantically tag contemporary documents. For this purpose, the Wikipedia and OpenCyc knowledge bases, which are known to contain up to date instances and reliable metadata about them, are selected. Experiments show that the accuracy of the mapping between WordNet and Wikipedia is 84% for the most relevant concept name and 90% for the appropriate sense.
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.