Data Mining - Concepts and Techniques.

We will review some of the major results in random graphs and some of the more challenging open problems. We will cover algorithmic and structural questions. We will touch on newer models, including those related to the WWW.

Random graphs

“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

Social Network Analysis

Emotion and Adaptation

In this study, a novel multi-classifier ensemble method based on dynamic weights is proposed to reduce the interference of unreliable decision information and improve the accuracy of fusion decision. The algorithm defines decision credibility to describe the real-time importance of the classifier to the current target, combines this credibility with the reliability calculated by the classifier on the training data set and dynamically assigns the fusion weight to the classifier. Compared with other methods, the contribution of different classifiers to fusion decision in acquiring weights is fully evaluated in consideration of the capability of the classifier to not only identify different sample regions but also output decision information when identifying specific targets. Experimental results on public face databases show that the proposed method can obtain higher classification accuracy than that of single classifier and some popular fusion algorithms. The feasibility and effectiveness of the proposed method are verified.

/pdf/multi-classifier-ensemble-based-on-dynamic-weights-1heuftxhc9.pdf

Multi-classifier ensemble based on dynamic weights

A New Question Answering System for Chinese Restricted Domain(Language, Human Communication II)

The research on machine translation has been lasting for many years, and now this research field is increasingly a thoroughly refined. In practical machine translation system, the processing of a simple and short Chinese sentence has somewhat good results. However, but the complex long Chinese sentence translation still has difficulties. In this paper a new hierarchical approach processing for Chinese complex long sentence through analysis of Chinese punctuation, conjunctive words and syntax function is proposed. The method synthetically uses semantic characteristic of source Chinese sentence, which includes grammatical features, the length of the source sentence, punctuation and functional words. First phase is conjunctive words for simplified segmentation, and then the syntax analysis, in order to process complex long sentence by the multi-hierarchical approach. A long sentence is divided into several parts; every part gets the correct translation of the result separately, and then is combined by the comprehensive approach to gain the complex long sentence translation result. Experiments show that our approach can significantly reduce the time consumption and numbers of ambiguity, and also improve the accuracy and readability when parsing Chinese complex long sentence.

Chinese complex long sentences processing method for Chinese-Japanese machine translation

Development of information technology has increased the chances of interaction between human and computer. Understanding not only semantic information but also underlying emotion will make communication with computer more natural and congenial. As preliminary research for emotion estimation from multilingual text, we focused on emotion expressions of Japanese and English and created a corpus with about 1,200 bilingual sentences and tagged emotions on words and sentences. Based on the corpus, we statistically analyzed the characteristics of how emotions were expressed in each language and also the relationships between sentence emotion and word emotion in the sentence. The results show the characteristics of emotion expressions in Japanese and English dialogue sentences.

Corpus-based Analysis of Japanese-English Emotional Expressions

Falls are the second leading cause of injury deaths worldwide, inducing over $0.6$ million accidental deaths per year. Among various prevention strategies, fall-related research has been prioritized. However, conventional fall detection solutions relying on computer vision or wearable sensors embody several inherent limitations such as scalability, coverage, and privacy issues. To this end, we present FallSense, a transparent and real-time fall sensing system driven by wireless channel data. FallSense is built on a Dynamic Template Matching (DTM) algorithm, which can start with a light training set and keep updating on usage. FallSense has been realized on commodity WiFi devices and evaluated in real environments. Experimental results show that FallSense outperforms another state-of-the-art approach WiFall in terms of detection precision, false alarm rate and complexity.

Fuji Ren

Papers

Multi-classifier ensemble based on dynamic weights

A New Question Answering System for Chinese Restricted Domain(Language, Human Communication II)

Chinese complex long sentences processing method for Chinese-Japanese machine translation

Corpus-based Analysis of Japanese-English Emotional Expressions

Your WiFi Knows You Fall: A Channel Data-driven Device-free Fall Sensing System