Syohei Ishizu

Rough ontology: extension of ontologies by rough sets

Ontology is widely used in the areas of knowledge engineering, web based data mining, etc. In rough set theory, accuracy of approximation of set and a concept of granularity are introduced. Rough set theory is very useful to define dependency among attributes and extract decision rules from the set. One of our main aims of this paper is to propose a concept of rough ontology. A concept of rough ontology is extended concept of rough set, and it enables us to use flexible information system in the form of ontology. And rough ontology is useful to introduce concepts rough set theory in to ontology. In this paper we formulate a concept of rough ontology, which is extended concept of rough set theory. We define upper and lower approximation, accuracy of approximation of preference, concept of granularity of preference. And we also show the property of rough ontology.

Evaluation of Total Quality Management Using CSR Company Reports

In recent years, serious quality accidents and quality troubles such as recalls have occurred frequently, and Total Quality Management (TQM) is important as effective management to prevent quality troubles beforehand. There is also a Quality Management Level Research [6] (TQM research) that is jointly implemented by the Union of Japanese Scientists and Engineers and the Nikkei. In the TQM survey, the rankings on the six criteria of quality management and comprehensive rankings have been announced. However, concrete TQM activities have not been announced. Meanwhile, Corporate Social Responsibility (CSR) report has published abundant descriptions about the role of customers, employees, society and management who are stakeholders of companies. In this research, we aim to evaluate and extract the characteristics of the company’s quality management activities according to the six criteria of the TQM survey using corporate CSR reports.

Development of a price promotion model for online store selection

There are many customer concerns related to online shopping, such as the inability to view actual products and the possibility of dishonesty. Online shopping nevertheless has the advantage of generally low prices. Effective price promotion that considers both customer concerns and price advantage is important for online stores. We developed a store selection model for both online stores and brick-and-mortar stores. We also conducted a survey to test the store selection model. Finally, we propose an effective price promotion method for each type of store.

Extracting Important Knowledge from Multiple Markets Using Transfer Learning

The aim of this study is to extract a customer’s needs from their reviews of an electronic commerce (EC) site using transfer learning. Transfer learning involves retaining and applying the knowledge learned from one or more tasks to efficiently develop an effective hypothesis for a new task. Recently, with the spread of EC sites, customer reviews have become a beneficial information source, as they include customers’ opinions or product reputations, and can attract attention. However, this information is too huge to browse conveniently. Moreover, to develop new products with a competitive advantage, it is necessary to incorporate customers’ opinions. Therefore, it is necessary to extract the customers’ opinions from the enormous amount of customer reviews.

A Quality Table-Based Method for Sentiment Expression Word Identification in Japanese

Identifying and summarizing opinions from online reviews is a valuable and challenging task and aspect-level sentiment analysis is a research-based approach to this task. Sentiment expression word identification is important sentiment identification task since many unique expression words appear in each entity domain and it is confirmed that text data from the internet has many collateral expressions. Generally, syntax-based model is applied to sentiment expression word identification method. Syntax-based model can consider low frequency word; however, we need to consider many syntax relations and that may be not practical. Therefore, it is difficult to identify sentiment expression words with syntax-based model. This paper proposes quality table-based method for sentiment expression word identification. The method identifies sentiment expression words with supervised learning. The training set is created with both seed expression-aspect and word-aspect deployment based on characteristic of quality table’s relation. This paper proposes a non-syntax and relation-based model in order to solve syntax-based models’ problems. This paper carries out an experimental test, demonstrates how many unique SEWs are extracted, and verifies the coverage of SEW with annotated text.

Knowledge Extraction About Brand Image Using Information Retrieval Method

The purpose of this study is to extract characteristic words as the information of expression pertaining to a brand’s image, from the language resources that have accumulated by users on Twitter. In this study, we analyzed Twitter data related to brands extracted the characteristic representations by using Okapi BM25, which is a ranking function that has been recently introduced in information retrieval. To confirm the validity of our approach, we conduct comparative experiments on the Twitter data of several Japanese automobile brands using BM25 and TF-IDF. By using the BM25, the extraction of keywords that are meaningful and buried in high frequency terms can be expected.

Development of Brand Selection Model Considering Customer Service

In many companies, customer service becomes one of the critical factors of the brand evaluation. It is important for the companies to know the customer’s utility functions about the customer service and repurchase of the products considering customer service. One of our main aims of this study is to develop brand selection model which considering utility of the customer service in order to propose promotion method for customer service of the company. In this study, we add some types of the customer services to utility in the models, and develop brand selection models. By the use of the questionnaire survey for the real companies, we confirm the adaptability of the proposed model, and we show the importance of the customer service.

Visualization of anomaly data using peculiarity detection on learning vector quantization

The purpose of this research is to develop the control chart robust for complex multidimensional data. In this study, we propose the methodology of anomaly data visualization and detection using hybrid model of Learning Vector Quantization (LVQ) and Peculiarity Factor (PF). LVQ is neural network model which uses supervised learning algorithm. It is useful to classification of multidimensional data with nonlinearity and multi-collinearity. PF is a criterion for evaluating peculiarity and is widely used for outlier detection. In the proposing method, PF of input data is calculated using the weight vector of LVQ. The anomaly data assigned to the class of the normal data was able to be displayed as an outlier on the control chart by calculation of PF on LVQ. The proposed model realized the robust discernment and visualization of the anomaly data that have complex distribution by small computational complexity.

A method for developing quality function deployment ontology

It is important to provide developed products in accord with customer needs to the market. We usually use QFD (Quality Function Deployment) to assure the quality fit for the customer needs. We can check the completeness of the qualities which are necessary to realize the customer needs by QFD, and compute importance of qualities in terms of QA. Supporting tools for QFD make quality table and compute importance of qualities. Moreover, QFD tools with ontologies are developed to treat hierarchy of qualities. However in real QFD, we cannot deploy without technical knowledge of design and manufacturing engineers. The relationships among the qualities sometimes change by product mechanisms of technical condition and we must consider various conditions when we perform horizontal deployment. Thus, we need a supporting tool that can represent logical restrictions of incorporating product mechanism. Our main aim of this paper is to propose a methodology of QFD based on logical restrictions and propose a supporting tool for QFD.

Software Testing Method Considering the Importance of Factor Combinations in Pair-Wise Testing

Software testing bears a burden of software development, increasing its time and cost. The bugs appearing due to the combination of two factors are well known in the system test phase. The current system testing methods represented by pair-wise tests or orthogonal arrays tests generate test sets by the forms of factors and values. In this study we extract two problems in the system test phase, and propose a solution to solve the problems. The first type of the problems is a survival bugs by the combination of factors among test sets. The second type of the problems is a duplication of factors by extra test case in the test set. We propose a solution which considers combinations of important factors for these two problems.