Collaborative filtering recommender system suffers from data sparsity problem due to its reliance on numerical ratings to provide recommendations to users. This problem makes it difficult for the system to compute accurate similar neighbours for the items and provide good quality recommendations. Existing methods fail to pre-process the missing ratings of the new items and to predict cold items to the active users which lead to poor quality recommendations. In this work, a sparsity reduction method is presented to improve the quality of recommendations. The method utilises Bi-Separated clustering algorithm to cluster the ratings matrix simultaneously into users and items bi-clusters based on ratings classification. It also employs Bi-Mean Imputation algorithm to fill the missing ratings in the bi-clusters using the estimated means. The method then performs the traditional collaborative filtering process on the new rating matrix for cold items prediction. The experimental results demonstrated that compared to the existing method, the proposed BiSCBiMI improves density of the rating matrix by 5.75%, 10.73% and 7.35% as well as Mean Absolute Error (MAE) of the new items prediction for all of the considered datasets. The results indicated that, the proposed approaches are effective in reducing the data sparsity problem as well as items prediction, which in turn returns good quality recommendations.

Java Cases and Ontology Libraries Integration for Building Reasoning Infrastructures (jCOLIBRI) is a framework which makes the development of Textual Case-Based Reasoning (CBR) applications easier by providing the preprocessing of text methods, textual similarity methods and appropriate representation for textual cases which are the major techniques needed in any CBR systems. In this paper, a Mobile Phone Diagnosis Support System is presented as an extension to jCOLIBRI which accepts a problem and reasons with cases to provide a solution related to a new given problem. Experimental evaluation using some set of problems shows that the developed system predicts the solution that is relatively closer to the user given mobile phone problem. The solution also provide the user valuable advise on how to go about solving the new problem.

Recommender systems (Rs) are widely used to provide recommendations for a collection of items or products that may be of interest to user or a group of users. Because of its superior performance, Content-Based Filtering (CBF) is one of the approaches that are commonly utilized in real-world Rs using Time-Frequency and Inverse Document Frequency (TF-IDF) to calculate document similarities. However, it computes document similarity directly in the word-count space. We propose a user-based collaborative filtering (UBCF) method to solve the problem of limited in content analysis which leads to a low prediction rate for large vocabulary. In this study, we present an algorithm that utilises Euclidean distance similarity function, to solve the identified problem. The performance of the proposed scheme was evaluated against the benchmark scheme using different performance metrics. The proposed scheme was implemented and an experimentally tested by using the benchmark datasets (Amazon review datasets). The results revealed that, the proposed scheme achieved better performance than the existing recommender system in terms of Root Mean Square Error (RMSE) which reduces the errors by 29% and also increase the Precision and Recall by 51.4%, and 55.8% respectively in the 1 million datasets.

Even though there are various source code plagiarism detection approaches, most of them are only concerned with lexical similarities attack with an assumption that plagiarism is only conducted by students who are not proficient in programming. However, plagiarism is often conducted not only due to student incapability but also because of bad time management. Thus, semantic similarity attacks should be detected and evaluated. This research proposes a source code semantic similarity detection approach that can detect most source code similarities by representing the source code into an Abstract Syntax Tree (AST) and evaluating similarity using a Siamese neural network. Since AST is a language-dependent feature, the SOCO dataset is selected which consists of C++ program codes. Based on the evaluation, it can be concluded that our approach is more effective than most of the existing systems for detecting source code plagiarism. The proposed strategy was implemented and an experimental study based on the AI-SOCO dataset revealed that the proposed similarity measure achieved better performance for the recommendation system in terms of precision, recall, and f1 score by 15%, 10%, and 22% respectively in the 100,000 datasets. In the future, it is suggested that the system can be improved by detecting inter-language source code similarity.

Textual Case-Based Reasoning as a problem solving approach allows knowledge source to be integrated with a view to improving the effectiveness of the system during retrieval. The earlier proposed Textual Case-based System depends on statistical similarity alone and most of the time does not retrieve the solution to the problem even if it exists. In this paper, the WordNet is being integrated to the developed Textual Case-Based Mobile Phone Diagnosis Support system in order to take the synonyms similarity of the problem terms into account while diagnosing a given problem. Thus, the integration will makes the system not to depend on statistical similarity alone but rather take synonyms similarity of the problem term into consideration. The result of the experimental evaluation using some set of problems has demonstrated that retrieval by incorporating WordNet works better since it diagnosed 95% of the problems with relevant solutions than the retrieval without WordNet which diagnosed 75% of the problems with relevant solutions.