AUTHOR & AFFILIATION
DR . FANDY YONK
Otto-von-guericke University, dept. of computer science, magd burg, Germany
VOL. 02 ISSUE 04
This essay establishes CIIT, the foremost hybrid research paper recommender system along with a powerful substitute to presently used academic search engines. CIIT develops the look of the typically implemented keyword-based search through uniting it by citation analysis, author analysis, implicit ratings, source analysis, explicit ratings and also, pioneering and yet idle methods similar to the ‘Distance Similarity Index’ (DSI) as well as the ‘In-text Impact Factor’ (ItIF). Rather than searching just keywords, a user may give whole documents, as well as reference lists as input as well as make implicit and explicit ratings to develop recommendations. With a quotation, author along with source analysis, similar and associated documents are effortlessly determinable. Every one of these methods is managed with an accessible GUI.
DS I, Recommendation, Recommender Systems, Research paper.
Numerous scientists think the search for connected work is a tremendously lengthy division of their responsibilities. The extent of time taken is partially caused by the rising amount of publications, which develops exponentially at an annual rate of 3.7 % . The strength of presently implemented academic search engines resides in searching documents including precise keywords. Because of synonyms and imprecise classifications, this approach brings in practice, regularly rambling results.
In this essay, we focus on CIIT1, a hybrid recommender system, which implements both content-based as well as collaborative-based methods. We consider that this approach has the possibility to lessen the difficulty of searching related research papers. Rather than exclusively deriving text mining, CIIT combines citation analysis, explicit ratings, author analysis, implicit ratings, as well as source analysis toward a recommender system by an accessible GUI. Currently, CIIT is inside the development phase and open for collaboration.
The primary part of this essay gives an impression of associated work as well as a discussion of the benefits and difficulties of accessible approaches. The major part applies CIIT and converses with the technologies used. The spotlight lies on top of a hybrid recommender method, which joins content-based along with collaborative-based methods. It explains that countless of the inconveniences of presented systems become outdated by combining well-known concepts with new ones. The last section of the paper provides insights into the procedure of the software by illustrating its functionality by screenshots.
- May, M. 1997. The Scientific Wealth of Nations, Science, vol. 275, no. 5301, pp. 793-796.
- Torres, R. McNee, S. M. Abel, M. Konstan, J. A.. and Riedl, J. 2004. Enhancing Digital Libraries with TechLens, Proceedings of JCDL’04, 228-236.
- Pennock, D. M. Horvitz, E. Lawrence, S. and Giles, L. C. 2000. Collaborative Filtering by Personality Diagnosis: A Hybrid Memory- and Model-Based Approach, in Proceedings of the Sixteenth Conference on Uncertainty in Artificial Intelligence (San Francisco).
- Middleton, S.E. Shadbolt, N. R., and De Roure, D. C. 2004. Ontological User Profiling in Recommender Systems, ACM Transactions on Information Systems (TOIS), 22, no.1, pp. 54-88.
- Fano, R. M. 1956. Information theory and the retrieval of recorded information, in Documentation in Action, Shera, J. H.
- Kent, A. Perry, J. W. (Edts), New York: Reinhold Publ. Co., pp. 238–244.
AUTHOR & AFFILIATION
PROF. DR . FANDY YONK
OTTO-VON-GUERICKE UNIVERSITY, DEPT. OF COMPUTER SCIENCE, MAGDEBURG, GERMANY