Eric W. Cooper

Borderline over-sampling for imbalanced data classification

Traditional classification algorithms usually provide poor accuracy on the prediction of the minority class of imbalanced data sets. This paper proposes a new method for dealing with imbalanced data sets by over-sampling the borderline minority class instances. A Support Vector Machine (SVM) classifier is then trained to predict future instances. Compared with other over-sampling methods, the proposed method focuses only on the minority class instances residing along the decision boundary, due to the fact that this region is the most crucial for establishing the decision boundary. Furthermore, the artificial minority instances are generated in such a way that the regions of the minority class with fewer majority class instances would be expanded by extrapolation, otherwise the current boundary of the minority class would be consolidated by interpolation. Experimental results show that the proposed method achieves a better performance than other over-sampling methods.

Online learning from imbalanced data streams

Learning from imbalanced data has conventionally been conducted on stationary data sets. Recently, there have been several methods proposed for mining imbalanced data streams, in which training data is read in consecutive data chunks. Each data chunk is considered as a conventional imbalanced data set, making it easy to apply sampling methods to balance data chunks. However, one drawback of chunk-based learning methods is that the update of classification models is delayed until a full data chunk is received. Therefore, this paper proposes a new method for online learning from imbalanced data streams, which uses naive Bayes as the base learner. To deal with the problem of class imbalance, a new training instance from the minority class is always involved in learning, but one from the majority class is only used with a small probability. In effect, this method corresponds to an under-sampling technique on imbalanced data streams. We show the effectiveness of the proposed online learning method on ten UCI data sets of various domains. Problems in the performance of naive Bayes on imbalanced data sets are also discussed.

Multi-command Tactile Brain Computer Interface: A Feasibility Study

The study presented explores the extent to which tactile stimuli delivered to the ten digits of a BCI-naive subject can serve as a platform for a brain computer interface BCI that could be used in an interactive application such as robotic vehicle operation. The ten fingertips are used to evoke somatosensory brain responses, thus defining a tactile brain computer interface tBCI. Experimental results on subjects performing online real-time tBCI, using stimuli with a moderately fast inter-stimulus-interval ISI, provide a validation of the tBCI prototype, while the feasibility of the concept is illuminated through information-transfer rates obtained through the case study.

Hybrid Kansei-SOM model using risk management and company assessment for stock trading

Risk management and stock assessment are key methods for stock trading decisions. In this paper, we present a new stock trading method using Kansei evaluation integrated with a Self-Organizing Map model for improvement of a stock trading system. The proposed approach aims to aggregate multiple expert decisions, achieve the greatest investment returns, and reduce losses by dealing with complex situations in dynamic market environments, such as downward, upward, steady market trends, and other uncertain conditions. Kansei evaluation and fuzzy evaluation models are applied to quantify trader sensibilities about stock trading, market conditions, and stock market factors with uncertain risks. In Kansei evaluation, group psychology and sensibility of traders are quantified that represent in fuzzy weights. Kansei and stock-market data sets are visualized by SOM, together with aggregate expert preferences in order to find potential companies, matching with trading strategies at the right time and eliminating risky stocks. The proposed approach has been tested and performed well in daily stock trading on the HOSE, HNX (Vietnam), NYSE and NASDAQ (US) stock markets. The experiments through case studies show that the new approach, applying Kansei evaluation enhances the capability of investment returns and reduce losses. The experimental results also show that the proposed approach performs better than other current methods to deal with various market conditions.

A comparative study on sampling techniques for handling class imbalance in streaming data

Sampling is the most popular approach for handling the class imbalance problem in training data. A number of studies have recently adapted sampling techniques for dynamic learning settings in which the training set is not fixed, but gradually grows over time. This paper presents an empirical study to compare over-sampling and under-sampling techniques in the context of data streaming. Experimental results show that under-sampling performs better than over-sampling at smaller training set sizes. All sampling techniques, however, are comparable when the training set becomes larger. This study also suggests that a multiple random under-sampling (MRUS) technique should be a good choice for applications with imbalanced and streaming data, because MRUS is the most effective while still keeping a high speed.

A Proposed Model of Diagnosis and Prescription in Oriental Medicine Using RBF Neural Networks

In this paper, we present a computing model for diagnosis and prescription in oriental medicine. Inputs to the model are severities of symptoms observed on patients and outputs from the model are a diagnosis of disease states and treatment herbal prescriptions. First, having used rule inference with a Gaussian distribution, the most serious disease state in which the patient appears to be infected is determined. Next, an herbal prescription written in suitable herbs with reasonable amounts for treating the infected disease state is given by RBF neural networks. Finally, we show some experiments and their evaluations, and then describe our future works.

Kansei and colour harmony models for townscape evaluation

Abstract Townscape colours have been a main factor in urban development. For townscape colours, keeping colour harmony within the environment is a common goal. Expressing characteristics and impressions of the town in townscape colours are other meaningful goals. The colour planning support system proposed here is intended to improve townscapes. The system offers some colour combination proposals based on three elements: colour harmony, impressions of the townscape, and cost for the change of colours. The objective of the present paper is to construct the colour harmony and Kansei evaluation models that evaluate colour combinations in the colour planning support system. The colour harmony equations by Moon and Spencer are employed for the construction of the colour harmony model. The Kansei model, which quantifies the impressions of the townscape, is constructed from the approach of Kansei engineering with neural networks. After the construction, evaluation experiments are conducted for 20 subjects to test the performance of both models. The results of the tests show sufficient correlation between model output and subject response for each model.

A townscape evaluation system based on Kansei and colour harmony models

Townscape colours have been a main factor in urban-development problems. In a townscape, keeping harmony within the environment is a common goal. But expressing individuality and impressions of the town are also meaningful goals. We proposed a colour planning support system to improve townscapes. The system finds propositional colour combinations based on three elements: impressions of the townscape, colour harmony, and cost. We mainly discuss the evaluation methods in Kansei and colour harmony models. The colour harmony model evaluates the townscape colours from the approach of colour harmony. To construct the model, we adopted the colour harmony equations from those proposed by Moon and Spencer. After the construction, we conducted evaluation experiments for 20 subjects to test the model performance. The testing showed sufficient correlation between model output and subject response for implementation of the harmony model in the townscape evaluation system.

A computational model to relay emotions with tactile stimuli

Haptic sensations have a long history of semantic relationship with various cognitive states and communication experiences. In developing multimedia interfaces, engineers would benefit greatly from knowing how sense of touch relates to human emotions. An important step in understanding this relationship is to study possible links between linguistic representations of the major haptic and emotional groups. This paper describes an attempt to reveal the semantic associations between basic categories of emotion and primary haptic sensations apparently existing in a culturally homogeneous group. A computational model is proposed to relay emotional experiences assessed on scales of “pleasure-unpleasure” and ”anxiety-boredom” with a standard haptic display, such as PHANToM. The model is defined in basic terms of physics, and can be used in virtually any application supporting haptic environments.

Townscape Color Planning System Using an Evolutionary Algorithm and Kansei Evaluations

Townscape colours have been a main issue in urban-development. For townscape colours, keeping colour harmony within the environment is a common goal. Expressing characteristics and impressions of the town in townscape colours are other meaningful goals. This paper describes the colour planning support system intended to improve town-scapes. The system offers some colour combination proposals based on three elements: colour harmony, impressions of the townscape, and cost for the change of colours. First, we develop evaluation models to quantify colour harmony and impression of the townscape from the approach of Kansei engineering. Next, the system is constructed using an evolutionary algorithm and the two evaluation models. After the construction, performance tests are conducted. The results show that our system achieved sufficient ability to propose appropriate colour combinations with minimum colour changes.