Tien-Chin Wang

Developing a fuzzy TOPSIS approach based on subjective weights and objective weights

Multiple criteria decision making (MCDM) is widely used in ranking one or more alternatives from a set of available alternatives with respect to multiple criteria. Inspired by MCDM to systematically evaluate alternatives under various criteria, we propose a new fuzzy TOPSIS for evaluating alternatives by integrating using subjective and objective weights. Most MCDM approaches consider only decision makers subjective weights. However, the end-user attitude can be a key factor. We propose a novel approach that involves end-user into the whole decision making process. In this proposed approach, the subjective weights assigned by decision makers (DM) are normalized into a comparable scale. In addition, we also adopt end-user ratings as an objective weight based on Shannons entropy theory. A closeness coefficient is defined to determine the ranking order of alternatives by calculating the distances to both ideal and negative-ideal solutions. A case study is performed showing how the propose method can be used for a software outsourcing problem. With our method, we provide decision makers more information to make more subtle decisions.

Multi-level fuzzy mining with multiple minimum supports

Finding association rules in transaction databases is most commonly seen in data mining. In real applications, different items may have different support criteria to judge their importance, taxonomic relationships among items may appear, and data may have quantitative values. This paper thus proposes a fuzzy multiple-level mining algorithm for extracting knowledge implicit in quantitative transactions with multiple minimum supports of items. Items may have different minimum supports and the maximum-itemset minimum-taxonomy support constraint is adopted to discover the large itemsets. Under the constraint, the characteristic of downward-closure is kept, such that the original apriori algorithm can be easily extended to find fuzzy large itemsets. The proposed algorithm adopts a top-down progressively deepening approach to derive large itemsets. It can also discover cross-level fuzzy association rules under the maximum-itemset minimum-taxonomy support constraint. An example is also given to demonstrate that the proposed mining algorithm can derive the multiple-level association rules under multiple item supports in a simple and effective way.

Forecasting the probability of successful knowledge management by consistent fuzzy preference relations

This paper presents an analytic hierarchy prediction model based on the consistent fuzzy preference relations to help the organizations become aware of the essential factors affecting the success of Knowledge Management (KM) implementation, forecasting the possibility of successful KM project, as well as identifying the actions necessary before initiating KM. Pairwise comparisons are utilized to obtain the priority weights of influential factors and the ratings of two possible outcome (success and failure). The subjectivity and vagueness within the prediction process are dealt with using linguistic variables quantified in an interval scale [0,1]. By multiplying the weights of influential factors and the ratings of possible outcome, predicted success/failure values are determined to enable organizations to decide whether to initiate knowledge management, inhibit adoption or take remedial actions to enhance the possibility of successful KM project. This proposed approach is demonstrated with a real case study involving seven major influential factors assessed by eleven evaluators solicited from a semiconductor engineering incorporation located in Taiwan.

Accurately predicting the success of B2B e-commerce in small and medium enterprises

Since implementing B2B e-commerce in small and medium enterprises (SMEs) is a long-term commitment and such enterprises are more limited in terms of resources than large enterprises, the predicted value of successful implementation is extremely useful in deciding whether to initiate B2B e-commerce. This investigation establishes an analytical hierarchy framework to help SMEs predicting implementation success as well as identifying the actions necessary before implementing B2B e-commerce to increase e-commerce initiative feasibility. The consistent fuzzy preference relation is used to improve decision-making consistency and effectiveness. A case study involving six influences solicited from a Taiwanese steel company is used to illustrate the feasibility and effectiveness of the proposed approach. The analytical results show that the three most influential factors are management support, industry characteristics and government policies; meanwhile, the three least influential factors are organizational culture, IT integration and firm size.

Applying Fuzzy PROMETHEE Method for Evaluating IS Outsourcing Suppliers

The demand for outsourcing in the information systems (IS) field has become a part of basic corporate strategy and has experienced a considerable growth in recent years. The process of IS outsourcing is always involved in a variety of situations and decision problems under fuzzy environment. The selection of appropriate outsourcing partners thus is one of the most important decision issues for organizations. The aim of this paper is to present the fuzzy preference ranking organization method for Enrichment Evaluation (fuzzy PROMETHEE) method to evaluate four potential suppliers based on seven criteria and four decision makers by using a realistic case study. The results of the rankings provide the reference to assist decision makers or organizations to improve the efficiency of their IS outsourcing decision processes come up with the best solutions.

Fuzzy multi-criteria selection among transportation companies with fuzzy linguistic preference relations

Selecting an appropriate transportation company is an important decision for an effective supply chain. This study attempts to solve transportation company selection problems initially addressed by Kulak and Kahraman in 2005, by adopting two methods. Consistent fuzzy preference relations presented by Herrera-Viedma et al. and fuzzy linguistic preference relations (fuzzy LinPreRa) developed by Wang and Chen in 2008. Analytical results indicate that both methods produce a consistent decision results from only n-1 pairwise comparisons. However, assigning linguistic variables to judgments is simpler and more intuitive than fixed value judgments. Therefore, the fuzzy LinPreRa is more suitable and efficient for providing rankings of transportation companies for making decisions.

Mining Fuzzy Multiple-level Association Rules under Multiple Minimum Supports

Finding association rules in transaction databases is most commonly seen in data mining. In real applications, different items may have different support criteria to judge their importance, taxonomic relationships among items may appear, and data may have quantitative values. This paper thus proposes a fuzzy multiple-level mining algorithm for extracting knowledge implicit in quantitative transactions with multiple minimum supports of items. Items may have different minimum supports and the maximum-itemset minimum-taxonomy support constraint is adopted to discover the large itemsets. Under the constraint, the characteristic of downward-closure is kept, such that the original Apriori algorithm can be easily extended to find fuzzy large itemsets. The proposed algorithm adopts a top-down progressively deepening approach to derive large itemsets. It can also discover cross-level fuzzy association rules under the maximum-itemset minimum-taxonomy support constraint. An example is also given to demonstrate that the proposed mining algorithm can derive the multiple-level association rules under multiple item supports in a simple and effective way.

Mining multiple-level association rules under the maximum constraint of multiple minimum supports

In this paper, we propose a multiple-level mining algorithm for discovering association rules from a transaction database with multiple supports of items. Items may have different minimum supports and taxonomic relationships, and the maximum-itemset minimum-taxonomy support constraint is adopted in finding large itemsets. That is, the minimum support for an itemset is set as the maximum of the minimum supports of the items contained in the itemset, while the minimum support of the item at a higher taxonomic concept is set as the minimum of the minimum supports of the items belonging to it. Under the constraint, the characteristic of downward-closure is kept, such that the original Apriori algorithm can easily be extended to find large itemsets. The proposed algorithm adopts a top-down progressively deepening approach to derive large itemsets. An example is also given to demonstrate that the proposed mining algorithm can proceed in a simple and effective way.

Mining Generalized Association Rules from a Different Perspective

In this paper, we introduce a mining algorithm from a different perspective for discovery of generalized association rules with multiple minimum supports. The perspective can be easily explained from the operations of union and intersection. Under the perspective, the characteristic of downward closure can be kept, such that the original Apriori algorithm can easily be extended to finding large itemsets. The proposed algorithm can thus meet the mining requirements of generating association rules under multiple minimum supports and managing taxonomic relationships among items. Experimental results show the effects of the parameters used in the proposed mining algorithm.