Time filter

Source Type

Ghazanfar M.A.,University of Engineering & Technology, Taxila | Prugel-Bennett A.,University of Southampton
Expert Systems with Applications | Year: 2014

Recommender systems apply data mining and machine learning techniques for filtering unseen information and can predict whether a user would like a given item. This paper focuses on gray-sheep users problem responsible for the increased error rate in collaborative filtering based recommender systems. This paper makes the following contributions: we show that (1) the presence of gray-sheep users can affect the performance - accuracy and coverage - of the collaborative filtering based algorithms, depending on the data sparsity and distribution; (2) gray-sheep users can be identified using clustering algorithms in offline fashion, where the similarity threshold to isolate these users from the rest of community can be found empirically. We propose various improved centroid selection approaches and distance measures for the K-means clustering algorithm; (3) content-based profile of gray-sheep users can be used for making accurate recommendations. We offer a hybrid recommendation algorithm to make reliable recommendations for gray-sheep users. To the best of our knowledge, this is the first attempt to propose a formal solution for gray-sheep users problem. By extensive experimental results on two different datasets (MovieLens and community of movie fans in the FilmTrust website), we showed that the proposed approach reduces the recommendation error rate for the gray-sheep users while maintaining reasonable computational performance.© 2013 Elsevier Ltd. All rights reserved.

Shehzad K.,University of Engineering & Technology, Taxila
IEEE Transactions on Knowledge and Data Engineering | Year: 2013

Pruning achieves the dual goal of reducing the complexity of the final hypothesis for improved comprehensibility, and improving its predictive accuracy by minimizing the overfitting due to noisy data. This paper presents a new hybrid pruning technique for rule induction, as well as an incremental postpruning technique based on a misclassification tolerance. Although both have been designed for RULES-7, the latter is also applicable to any rule induction algorithm in general. A thorough empirical evaluation reveals that the proposed techniques enable RULES-7 to outperform other state-of-the-art classification techniques. The improved classifier is also more accurate and up to two orders of magnitude faster than before. © 1989-2012 IEEE.

Ali H.M.,University of Engineering & Technology, Taxila | Briggs A.,Queen Mary, University of London
International Journal of Heat and Mass Transfer | Year: 2015

A simple semi-empirical correlation accounting for the combined effect of gravity and surface tension has been developed for condensation on horizontal pin-fin tubes. The model divides the heat transfer surface into five regions, i.e. two types of pin flank, two types of pin root and the pin tip. Data for three fluids (i.e. steam, ethylene glycol and R113) condensing on eleven tubes with different geometries were used in a minimization process to find three empirical constants in the final expression. The model gives good overall agreement (within ±20%) with the experimental data, as well as correctly predicting the dependence of heat-transfer enhancement on the various geometric parameters and fluid types. © 2014 Elsevier Ltd. All rights reserved.

Ali H.M.,University of Engineering & Technology, Taxila | Briggs A.,Queen Mary, University of London
Applied Thermal Engineering | Year: 2014

Retention angle measurements on 15 rectangular pin-fin tubes are made under static conditions (without condensation) using water, ethylene glycol and R-113. Condensate retention angles found to be considerably larger than the equivalent integral-fin tubes (i.e. with the same fin height, root diameter and longitudinal pin thickness and spacing) in a range of 5%-60%. An expression for condensate retention angle on pin-fin tubes was proposed and found to agree with the measured retention angles to within 15%. © 2013 Elsevier Ltd. All rights reserved.

Bailey C.G.,University of Manchester | Yaqub M.,University of Engineering & Technology, Taxila
Composite Structures | Year: 2012

An experimental study was carried out to investigate the seismic performance of post-heated circular reinforced concrete columns wrapped with glass or carbon fibre reinforced polymer jackets. Eight shear critical reinforced circular columns with a shear span-to-depth ratio of 2.5 were tested under a combined constant axial and cyclic lateral displacement history, simulating earthquake loading. The columns were tested in three groups, unheated, post-heated and post-heated repaired with either glass fibre reinforced polymer (GFRP) or carbon fibre reinforced polymer (CFRP). In terms of seismic performance the test results indicated that using GFRP or CFRP jackets significantly increased the shear capacity, ductility and energy dissipation of the post-heated damaged columns. However, the GFRP or CFRP did not increase the stiffness of the post-heated damaged columns. It was found that the unheated and post-heated damaged columns failed in a brittle shear mode while the mode of failure of posted-heated columns repaired with GFRP or CFRP was successfully shifted from a shear to a ductile flexural failure. © 2011 Elsevier Ltd.

Discover hidden collaborations