Samar Shilbayeh, Ph.D. (Ai And Machine Learning) Email & Phone Number

Dubai, Ae

Dubai, United Arab Emirates

Abu Dhabi

Abu Dhabi

Jordan

Saudi Arabia

Assistant professor at Princess Nourah University, computer science and information technology Collage

Manchester, United Kingdom

Cost sensitive Meta learning work This research developed a new meta-learning system for recommending cost-sensitive classification methods. The system is based on the idea of applying machine learning todiscover knowledge about the performance of different data mining algorithms. It includescomponents that repeatedly apply different classification methods on data sets and measuringtheir performance. The characteristics of the data sets, combined with the algorithm and… Show more Cost sensitive Meta learning work This research developed a new meta-learning system for recommending cost-sensitive classification methods. The system is based on the idea of applying machine learning todiscover knowledge about the performance of different data mining algorithms. It includescomponents that repeatedly apply different classification methods on data sets and measuringtheir performance. The characteristics of the data sets, combined with the algorithm and theperformance provide the training examples. A decision tree algorithm is applied on thetraining examples to induce the knowledge which can then be applied to recommendalgorithms for new data sets, and then active learning is used to automate the ability tochoose the most informative data set that should enter the learning process. This thesis makes contributions to both the fields of meta-learning, and cost sensitive learningin that it develops a new meta-learning approach for recommending cost-sensitive methods. Although, meta-learning is not new, the task of accelerating the learning process remains anopen problem, and the thesis develops a novel active learning strategy based on clusteringthat gives the learner the ability to choose which data to learn from and accordingly, speed upthe meta-learning process. Both the meta-learning system and use of active learning are implemented in the WEKAsystem and evaluated by applying them on different datasets and comparing the results withexisting studies available in the literature. The results show that the meta-learning systemdeveloped produces better results than METAL, a well-known meta-learning system and thatthe use of clustering and active learning has a positive effect on accelerating the metalearningprocess. Show less