Metric Methods in Data Mining

Data Mining Patterns ◽

10.4018/978-1-59904-162-9.ch001 ◽

2011 ◽

pp. 1-31 ◽

Cited By ~ 1

Author(s):

Dan A. Simovici

Keyword(s):

Data Mining ◽

Metric Space ◽

Training Data ◽

Future Research ◽

Open Problems ◽

Research Directions ◽

Data Mining Techniques ◽

Future Research Directions ◽

Major Data ◽

Geometric Study

This chapter presents data mining techniques that make use of metrics defined on the set of partitions of finite sets. Partitions are naturally associated with object attributes and major data mining problem such as classification, clustering, and data preparation benefit from an algebraic and geometric study of the metric space of partitions. The metrics we find most useful are derived from a generalization of the entropic metric. We discuss techniques that produce smaller classifiers, allow incremental clustering of categorical data and help user to better prepare training data for constructing classifiers. Finally, we discuss open problems and future research directions.

Download Full-text

Data Mining Analytics for Crime Security Investigation and Intrusion Detection

Advances in Data Mining and Database Management - Data Mining Trends and Applications in Criminal Science and Investigations ◽

10.4018/978-1-5225-0463-4.ch008 ◽

2016 ◽

pp. 212-244

Author(s):

Boutheina Fessi ◽

Yacine Djemaiel ◽

Noureddine Boudriga

Keyword(s):

Data Mining ◽

Dynamic Environments ◽

Future Research ◽

Cyber Crime ◽

Research Directions ◽

Data Mining Techniques ◽

Crime Investigation ◽

Security Investigation ◽

Future Research Directions ◽

Digital Investigation

This chapter provides a review about the usefulness of applying data mining techniques to detect intrusion within dynamic environments and its contribution in digital investigation. Numerous applications and models are described based on data mining analytics. The chapter addresses also different requirements that should be fulfilled to efficiently perform cyber-crime investigation based on data mining analytics. It states, at the end, future research directions related to cyber-crime investigation that could be investigated and presents new trends of data mining techniques that deal with big data to detect attacks.

Download Full-text

Data Mining Analytics for Crime Security Investigation and Intrusion Detection

Securing the Internet of Things ◽

10.4018/978-1-5225-9866-4.ch035 ◽

2020 ◽

pp. 700-725

Author(s):

Boutheina A. Fessi ◽

Yacine Djemaiel ◽

Noureddine Boudriga

Keyword(s):

Data Mining ◽

Dynamic Environments ◽

Future Research ◽

Cyber Crime ◽

Research Directions ◽

Data Mining Techniques ◽

Crime Investigation ◽

Security Investigation ◽

Future Research Directions ◽

Digital Investigation

This chapter provides a review about the usefulness of applying data mining techniques to detect intrusion within dynamic environments and its contribution in digital investigation. Numerous applications and models are described based on data mining analytics. The chapter addresses also different requirements that should be fulfilled to efficiently perform cyber-crime investigation based on data mining analytics. It states, at the end, future research directions related to cyber-crime investigation that could be investigated and presents new trends of data mining techniques that deal with big data to detect attacks.

Download Full-text

Explanation-Based Human Debugging of NLP Models: A Survey

Transactions of the Association for Computational Linguistics ◽

10.1162/tacl_a_00440 ◽

2021 ◽

Vol 9 ◽

pp. 1508-1528

Author(s):

Piyawat Lertvittayakumjorn ◽

Francesca Toni

Keyword(s):

Learning Model ◽

Training Data ◽

Three Dimensions ◽

Future Research ◽

Experimental Setting ◽

Open Problems ◽

Research Directions ◽

Machine Learning Model ◽

Future Research Directions ◽

Deep Learning Model

Abstract Debugging a machine learning model is hard since the bug usually involves the training data and the learning process. This becomes even harder for an opaque deep learning model if we have no clue about how the model actually works. In this survey, we review papers that exploit explanations to enable humans to give feedback and debug NLP models. We call this problem explanation-based human debugging (EBHD). In particular, we categorize and discuss existing work along three dimensions of EBHD (the bug context, the workflow, and the experimental setting), compile findings on how EBHD components affect the feedback providers, and highlight open problems that could be future research directions.

Download Full-text

MetaLearning with Graph Neural Networks

ACM SIGKDD Explorations Newsletter ◽

10.1145/3510374.3510379 ◽

2021 ◽

Vol 23 (2) ◽

pp. 13-22

Author(s):

Debmalya Mandal ◽

Sourav Medya ◽

Brian Uzzi ◽

Charu Aggarwal

Keyword(s):

Neural Networks ◽

Deep Neural Networks ◽

Future Research ◽

Open Problems ◽

Research Directions ◽

Graph Problems ◽

Meta Learning ◽

Comprehensive Survey ◽

Future Research Directions ◽

Graph Neural Networks

Graph Neural Networks (GNNs), a generalization of deep neural networks on graph data have been widely used in various domains, ranging from drug discovery to recommender systems. However, GNNs on such applications are limited when there are few available samples. Meta-learning has been an important framework to address the lack of samples in machine learning, and in recent years, researchers have started to apply meta-learning to GNNs. In this work, we provide a comprehensive survey of different metalearning approaches involving GNNs on various graph problems showing the power of using these two approaches together. We categorize the literature based on proposed architectures, shared representations, and applications. Finally, we discuss several exciting future research directions and open problems.

Download Full-text

Learning Analytics

Encyclopedia of Information Science and Technology, Fourth Edition ◽

10.4018/978-1-5225-2255-3.ch448 ◽

2018 ◽

pp. 5158-5168

Author(s):

Constanţa-Nicoleta Bodea ◽

Maria-Iuliana Dascalu ◽

Radu Ioan Mogos ◽

Stelian Stancu

Keyword(s):

Data Mining ◽

Learning Analytics ◽

Educational Data Mining ◽

Future Research ◽

Educational Systems ◽

Main Section ◽

Research Directions ◽

Data Intensive ◽

Academic Analytics ◽

Future Research Directions

Reinforcement of the technology-enhanced education transformed education into a data-intensive domain. As in many other data-intensive domains, the interest for data analysis through various analytics is growing. The article starts by defining LA, with relevant views on the literature. A discussion about the relationships between LA, educational data mining and academic analytics is included in the background section. In the main section of the article, the learning analytics, as an emerging trend in the educational systems is describe, by discussing the main issues, controversies, problems on this topic. Final part of the article presents the future research directions and the conclusion.

Download Full-text

Unit of Analysis in Digitally-Enabled Electronic Procurement Research

Digital Innovations for Customer Engagement, Management, and Organizational Improvement - Advances in Business Strategy and Competitive Advantage ◽

10.4018/978-1-7998-5171-4.ch005 ◽

2020 ◽

pp. 83-103

Author(s):

Md Mahbubur Rahim ◽

Maryam Jabberzadeh ◽

Nergiz Ilhan

Keyword(s):

Data Mining ◽

Cloud Computing ◽

Critical Issue ◽

Future Research ◽

Research Directions ◽

Research Issues ◽

Unit Of Analysis ◽

Electronic Procurement ◽

Future Research Directions ◽

Methodological Concern

E-procurement systems that have been in place for over a decade have begun incorporating digital tools like big data, cloud computing, internet of things, and data mining. Hence, there exists a rich literature on earlier e-procurement systems and advanced digitally-enabled e-procurement systems. Existing literature on these systems addresses many research issues (e.g., adoption) associated with e-procurement. However, one critical issue that has so far received no rigorous attention is about “unit of analysis,” a methodological concern of importance, for e-procurement research context. Hence, the aim of this chapter is twofold: 1) to discuss how the notion of “unit of analysis” has been conceptualised in the e-procurement literature and 2) to discuss how its use has been justified by e-procurement scholars to address the research issues under investigation. Finally, the chapter provides several interesting findings and outlines future research directions.

Download Full-text

Open Problems and Future Research Directions

Parameterized Complexity in the Polynomial Hierarchy - Lecture Notes in Computer Science ◽

10.1007/978-3-662-60670-4_16 ◽

2019 ◽

pp. 335-347

Author(s):

Ronald de Haan

Keyword(s):

Future Research ◽

Open Problems ◽

Research Directions ◽

Future Research Directions

Download Full-text

Computational and Data Mining Perspectives on HIV/AIDS in Big Data Era

10.4018/978-1-6684-3662-2.ch072 ◽

2022 ◽

pp. 1477-1503

Author(s):

Ali Al Mazari

Keyword(s):

Data Mining ◽

Big Data ◽

Data Analytics ◽

Big Data Analytics ◽

Healthcare Sector ◽

Future Research ◽

Research Directions ◽

Scientific Disciplines ◽

Future Research Directions ◽

Hiv Aids

HIV/AIDS big data analytics evolved as a potential initiative enabling the connection between three major scientific disciplines: (1) the HIV biology emergence and evolution; (2) the clinical and medical complex problems and practices associated with the infections and diseases; and (3) the computational methods for the mining of HIV/AIDS biological, medical, and clinical big data. This chapter provides a review on the computational and data mining perspectives on HIV/AIDS in big data era. The chapter focuses on the research opportunities in this domain, identifies the challenges facing the development of big data analytics in HIV/AIDS domain, and then highlights the future research directions of big data in the healthcare sector.

Download Full-text

Methods and Techniques of Data Mining

Encyclopedia of Information Science and Technology, Fifth Edition - Advances in Information Quality and Management ◽

10.4018/978-1-7998-3479-3.ch051 ◽

2021 ◽

pp. 749-767

Author(s):

Ana Funes ◽

Aristides Dasso

Keyword(s):

Artificial Intelligence ◽

Data Mining ◽

Research And Development ◽

Knowledge Acquisition ◽

Future Research ◽

Research Directions ◽

New Methods ◽

Methods And Techniques ◽

Future Research Directions

Nowadays, there is an increasing number of applications where artificial intelligence has fuelled the research and development of new methods, techniques, and tools related to knowledge acquisition and data mining. The development of data mining and other related disciplines has benefited from the existence of large volumes of data proceeding from the most diverse sources and domains. KDD process and methods of data mining allows for the discovery of knowledge in data that is hidden to humans, presenting this knowledge under different ways. In this chapter, the relation of data mining with other disciplines is analyzed, an overview of data mining tasks and methods is presented, and also a possible classification of them is given. Finally, a brief discussion on issues associated to the discipline and future research directions are also given.

Download Full-text