Lance Chun Che Fung

—Question answering over knowledge graphs (KGQA) seeks to automatically answer natural language questions by retrieving triples within the knowledge graph (KG). In the context of multi-hop KGQA, reasoning across multiple edges of the KG becomes crucial for obtaining answers. Existing methods align with either the path-searching-based mainstream, emphasizing structural KG analysis, or the subgraph-based mainstream, focusing on semantic KG embeddings. Both streams have two primary challenges: (1) KG incompleteness, where path searching or subgraph construction faces limitations in the absence of links between entities; (2) candidate answer selection, wherein most approaches employ pre-defined searching sizes or heuristics. Many recent studies incorporate Graph Convolutional Network (GCN) to encode KGs, yet they overlook the potential over-smoothing issue inherent in GCNs. The over-smoothing problem arises from the tendency of closely connected nodes to exhibit similar embeddings within the deep convolutional architecture of GCNs. To address these challenges, this paper proposes a two-stage framework named ComPath, leveraging insights from both mainstreams. ComPath utilizes GCN to tackle KG incompleteness and introduces a path analyser to mitigate the over-smoothing issue associated with GCN. Candidate answers are selected using semantic similarity. The ablation studies and comparative experiments on the three KGQA benchmark datasets shown that the proposed ComPath performed better than the other KGQAs.

Text clustering is an effective way that helps crime investigation through grouping of crime-related documents. This paper proposes a Memetic Algorithm Feature Selection (MAFS) approach to enhance the performance of document clustering algorithms used to partition crime reports and criminal news as well as some benchmark text datasets. Two clustering algorithms have been selected to demonstrate the effectiveness of the proposed MAFS method; they are the k-means and Spherical k-means (Spk). The reason behind using these clustering methods is to observe the performance of these algorithms before and after applying a hybrid FS that uses a Memetic scheme. The proposed MAFS method combines a Genetic Algorithm-based wrapper FS with the Relief-F filter. The performance evaluation was based on the clustering outcomes before and after applying the proposed MAFS method. The test results showed that the performance of both k-means and spk improved after the MAFS.

An overwhelming number of facet values causes difficulties in providing an efficient search filter in dynamic facet search. It requires effort and time from the searchers to examine the list in order to select their interested facets. Personalised facet selection provides a list of relevant facet which is related to the user's interests. However, personalisation may not be possible to determine a user's current interest from the user's profile or the user's history search only. In some cases, due to insufficient information to identify users' current interests, the need of associating community opinions with personal interests is necessary. This study aims to investigate the incorporation of a collaborative approach to personalise facet selection. Collaborative Filtering is employed to address the issue of limited profile information and the approach has been widely used in recommender systems. Experiments were conducted on a benchmark Movie dataset using user ratings as the representation of user preferences and evaluated by rating prediction accuracy and computational time. The results show that Collaborative Filtering should improve the performance of personalised facet selection.

Community question and answer (CQA) sites mostly involve knowledge-bases feeding into their automated question answering systems. This paper focuses on Stack Overflow which is an online community for developers to share knowledge in computer programming. The proposed framework consists of composing of a paired Q&A corpus, followed by building of a document model with the use of paragraph vector in distributed representation via the doc2vec method, then similarity ranking to fetch a matched answer to a given question. The model pairs the two so as to represent the semantic relevance between the questions and answers generated by the proposed method. The initial experimental results have shown the system is able to provide answers automatically and with a performance of 50% accuracy when compared to expert opinions.

This paper proposes algorithms for the extraction of features from candlestick patterns for technical analysis of share indices. The significant features consist of: the direction of candlestick, the gap between CLOSE and OPEN price of two candlesticks, the body level of current and previous candlesticks, and the length of the candlesticks. K-Means clustering approach is applied for solving the unclearly defined length of Upper Shadow, Body and Lower Shadow. The Thai SET index OHLC data from 1990 to 2017 are used as the experimental dataset. The results show the similarity between the candlestick chart from raw data and decoding data, which is applied by the proposed algorithms. The output result from the approach can be used as the input to other machine learning methods such as Artificial Neuron Networks, Reinforcement Learning, or Content Based Image Retrieval (CBIR).

With the wide increase of the volume of electronic documents, it becomes inevitable the need to invent more sophisticated machine learning methods to manage the issue. In this paper, a Memetic feature selection technique is proposed to improve the k-means and the spherical k-means clustering algorithms. The proposed Memetic feature selection technique combines the wrapper inductive method with the filter ranking method. The internal and external clustering evaluation measures are used to assess the resulted clusters. The test results showed that after using the proposed hybrid method, the resulted clusters were more accurate and more compacted in comparison to the clusters resulted from using the GA-selected feature or using the entire feature space.

A hybrid sampling technique is proposed by combining Complementary Fuzzy Support Vector Machine (CMTFSVM) and Synthetic Minority Oversampling Technique (SMOTE) for handling the imbalanced classification problem. The proposed technique uses an optimised membership function to enhance the classification performance and it is compared with three different classifiers. The experiments consisted of four standard benchmark datasets and one real world data of plant cells. The results revealed that implementing CMTFSVM followed by SMOTE provided better result over other FSVM classifiers for the benchmark datasets. Furthermore, it presented the best result on real world dataset with 0.9589 of G-mean and 0.9598 of AUC. It can be concluded that the proposed techniques work well with imbalanced benchmark and real world data.

Emotion classification and recognition from electroencephalogram (EEG) signals have been studied extensively due to its potential benefits such as entertainment and health care. Concerning classification, various techniques have been developed and applied. Support Vector Machines (SVMs) has been reported as the most used because of its accuracy. Nevertheless, although SVMs has satisfactory performance, it is unable to provide explanation of the relationships between a model's inputs and outputs. Specifically, it is desirable for a medical application for diagnosis to provide comprehensible rules. Consequently, SVM might not be suitable. In this study, SVM is treated as a black-box and then rules are extracted using the Classification And Regression Trees (CART) approach. A dataset from the Database for Emotion Analysis using Physiological Signals (DEAP) is used in this study. The experimental results show that although a classic SVM model has provided the best accuracy, a rule extraction model from SVM output by CART (SVM-CART) is better than a basic CART model. Therefore, the proposed SVM-CART approach is suitable for applications which need explanations and comprehensibility, such as medical applications.

A hybrid sampling technique is proposed by combining Complementary Fuzzy Support Vector Machine (CMTFSVM) and Synthetic Minority Over-sampling Technique (SMOTE) for handling imbalanced classification problem. The proposed technique is compared with three different classifiers. The optimize membership function is chosen to enhance the classification performance. The experiment uses the Glass5 dataset that has 22.78% of imbalanced ratio. The results revealed that by implementing CMTFSVM first and then applying SMOTE provided the best performance over the other methods with 0.9638 of G-mean and 0.9646 of AUC.

It is challenging to ensure security and to minimize economic impacts due to cyber-attacks because of the heavy reliance on ICT in different organizations and this paper presents an approach to estimate the cost of cyber security in public and private sector organizations. The paper also describes an approach for selecting the type of cyber security improvements to ensure that organizational goals are achieved. Different types of cyber-attacks and the subsequent impacts of these attacks are considered. A Value Analysis method is proposed to support the decision-making process by determining the priorities of deployment of various cyber security technologies. The proposed method is based on security costs related to and the losses due to attacks. Examples are provided in the paper to illustrate the proposed approach.