novel contributions of the doctoral dissertation

The thesis proposes the largest fuzzy distance concept, and is used as the basis to develop the decision tree algorithms based on the largest fuzzy disctance HA[r]

HUE UNIVERSITY

HUE UNIVERSITY OF SCIENCES

SOCIALIST REPUBLIC OF VIETNAM Independence - Freedom - Happiness

NOVEL CONTRIBUTIONS OF THE DOCTORAL DISSERTATION

1 GENERAL INFORMATION

Full name: Lê Văn Tường Lân

Topic: Data classification by fuzzy decision tree base on hedge algebra

Major: Computer Science Code: 62.48.01.01

Supervisors:

Assoc Prof Dr Nguyễn Mậu Hân,

Faculty of Infomation Technology, Hue University of Sciences

Dr Nguyễn Công Hào,

Department of Inspection and Legislation, Hue University

Institution: Hue University of Sciences, Vietnam

2 NOVEL CONTRIBUTIONS OF THE DISSERTATION

1 Propose a flexible model for decision tree learning from a practical training dataset and a method to extract a specific training set for the training process

2 Analyze and introduce the concept of extrinsic value in heterogeneous sample datasets Developing the algorithms to uniform these attributes that contain these values

3 Proposed MixC4.5 and FMixC4.5learning algorithms: based on the synthesis of the advantages and disadvantages of traditional algorithms (CART, C4.5, SLIQ, SPRINT)

4 Proposed a matching solution onfuzzy distance, and afuzzy clustering algorithmbased on a HAC4.5 fuzzy distance Develop a method to quantify the values of the non-uniform, unspecified Min-Max attributes of the training set

5 The thesis proposes the largest fuzzy distance concept, and is used as the basis to develop the decision tree algorithms based on the largest fuzzy disctance HAC4.5* to achieve the efficient classification and the simplicity for the user