150 likes | 251 Views
Multi-focal Learning and Its Application to Customer Service Support . Presenter : Tsai Tzung Ruei Authors : Yong Ge , Hui Xiong , Wenjun Zhou, Ramendra Sahoo,Xiaofeng Gao,Weili Wu. 國立雲林科技大學 National Yunlin University of Science and Technology. 2009.SIGKDD. Outline. Motivation
E N D
Multi-focal Learning and Its Application to Customer Service Support Presenter : Tsai TzungRuei Authors : Yong Ge, HuiXiong, Wenjun Zhou, RamendraSahoo,XiaofengGao,Weili Wu 國立雲林科技大學 National Yunlin University of Science and Technology 2009.SIGKDD
Outline • Motivation • Objective • Methodology • Experiments • Conclusion • Discussions • Comments
Motivation • All the problem descriptions for the same problem are provided by customers with diverse background and these problem descriptions can be quite different.
Objective • To formalize a multi-focal learning problem, where training data are partitioned into several different focal groups and the prediction model will be learned within each focal group. focal groups Problem descriptions Problem Solution
Methodology(2/3) • Focal Group Formation:CORRELATION • Focal Group Formation: ONTOLOGY
Methodology(3/3) • Risk Analysis of Multi-Focal Learning
Experiments(1/5) • Results on Problem Logs • Performance Comparisons • Results on Synthetic Data • Case Study
Experiments(2/5) • Results on Problem Logs
Experiments(3/5) • Performance Comparisons
Experiments(4/5) • Results on Synthetic Data
Experiments(5/5) • Case Study
Conclusion • The multi-focal learning allows the learning algorithms to mitigate the influence of the diversities inherent in training data, and thus leads to better learning performances. • Experimental results show that both CORRELATIONand ONTOLOGY have led to better learning performancesthan other focal-group formation methods, suchas the methods based on clustering and random-partition,while the learning performance by ONTOLOGY is lightlybetter than that by CORRELATION.
Discussions • For instance, let us consider a videosurveillance system. There are different types of moving objects,such as cars, bikes, and human beings. Those movingobjects have different sizes, speed, and moving capabilities.To better capture abnormal moving patterns, it is expectedto apply the multi-focal learning techniques to first groupmoving objects into different focal groups. The detectionof abnormal moving patterns can then be performed withindifferent focal groups.
Comments • Advantage • To boost the learning accuracies of existing learning algorithms, such as Support Vector Machines (SVMs), for classifying customer problems. • Drawback • Some mistakes • Application • Customer Service Support