International Journal of Information Technology & Decision Making

Data Envelopment Analysis (DEA) is a well-known nonparametric technique of efficiency evaluation which is actively used in many economic applications. However, DEA is not very well applicable when a sample consists of firms operating under drastically different conditions. We offer a new method of efficiency estimation in heterogeneous samples based on a sequential exclusion of alternatives and standard DEA approach. We show a connection between efficiency scores obtained via standard DEA model and the ones obtained via our algorithm. We also illustrate our model by evaluating 28 Russian universities and compare the results obtained by two techniques.

To identify series of residential burglaries, detecting linked crimes performed by the same constellations of criminals is necessary. Comparison of crime reports today is difficult as crime reports traditionally have been written as unstructured text and often lack a common information-basis. Based on a novel process for collecting structured crime scene information, the present study investigates the use of clustering algorithms to group similar crime reports based on combined crime characteristics from the structured form. Clustering quality is measured using Connectivity and Silhouette index (SI), stability using Jaccard index, and accuracy is measured using Rand index (RI) and a Series Rand index (SRI). The performance of clustering using combined characteristics was compared with spatial characteristic. The results suggest that the combined characteristics perform better or similar to the spatial characteristic. In terms of practical significance, the presented clustering approach is capable of clustering cases using a broader decision basis.

This paper presents a simulation study to assess the performance of the five known methods for converting ranks of several criteria into weights in multi-criteria decision-making. The five methods assessed for converting criteria ranks into weights are: rank- sum (RS) weights, rank reciprocal (RR) weights, rank order centroid (ROC) weights, geometric weights (GW), and variable-slope linear (VSL) weights. The methods are compared in terms of weight estimation accuracy considering different numbers of criteria and decision makers’ (MS) preference structures. Alternative preference structures are represented by different probability distributions of randomly generated criteria weights, namely the uniform, normal, and exponential distributions. Results of the simulation experiments indicate that no single method is consistently superior to all others. On average, RS is best for uniform weights, VSL is best for normal weights, and ROC is best for exponential weights. However, for any multi-criteria decision-making (MCDM) problem, the best method for converting criteria ranks into weights depends on both the number of criteria and the weight distribution.

Human preferences or attitudes towards risk should play a vital role in a decision making task with imperfect information and uncertain outcomes. We introduce a method to characterize human preferences and how they are integrated into the decision making process of a complex probability-based card bidding game. When assessing the preferences, a utility-to-preference (UP) function is devised for easier mapping between preferences and how much a player is willing to bid. Using the developed approach, we can better identify how different human preferences and their interaction affect the game outcomes. We focus on a highly addictive poker game that has become a multi-billion dollar internet business. The method was evaluated through data obtained from two decision makers (DMs) with different expertise. The integrated decision making process was designed and automated through Monte Carlo simulation. The results show that different preferences to the multi-attributes can lead to different profit outcomes. The results can further serve as a basis to identify vulnerable populations¹ for the socio-technical online bidding game.

Fuzzy sets, a scheme for handling nonstatistical vague concepts, provide a natural basis for the theory of possibility space. In this paper, on possibility space, a hierarchical generalization of the fuzzy hidden Markov chain (HFHMC) which is named as FHMC is proposed. For the proposed model, three problems which naturally arise in any kind of hidden Markov models (HMMs) are discussed. To solve these problems, generalized Baum–Welch and generalized Viterbi algorithms are formulated; further it is observed that the generalized Viterbi algorithm itself solves the first two problems namely the likelihood of a given observation sequence and finding the most likelihood state sequence, which exhibits that the time complexity involved in the computation of two problems reduces to a single problem. In order to ensure the ease of models use, the proposed model is applied to our institution website and simulation is performed to analyze the accessibility of the website among the users.

Customer expectation has been an important issue across different academic fields. Customer expectation management enables service providers to provide customers with suitable services in order to achieve high customer satisfaction especially in real-time dynamic service contexts. Understanding actual customer expectations is the primary step before managing customer expectations. However, to our knowledge, there is no research investigating how to measure real-time customer expectations during service delivery. Hence, this study proposes a customer expectation measurement mechanism and evaluates its feasibility and reliability through simulations. Simulation results show that the proposed mechanism offers researchers and service providers a feasible approach to measuring and managing real-time customer expectations at service encounters for building satisfactory customer experiences.

“Nobody ever got fired for buying IBM,” was a widely used cliché in the 1970s in the corporate IT (information technology) world. Since then, the traditional process of purchasing software has dramatically changed, challenged by the advent of open source software (OSS). Since its inception in the 1980s, OSS has matured, grown, and become one of the important driving forces of the enterprise ecosystem. However, it has also brought important IT security risks that are impacting the OSS IT adoption decision-making process. The recent Heartbleed bug demonstrated the grandeur of the issue. While much of the noise relates to the amplification of perceived risks by the popular mass media coverage, the effect is that many enterprises, mainly for risk reasons, have still chosen not to adopt OSS. We investigated “how do information security related characteristics of OSS affect the risk perception and adoption decision of OSS” by conducting an online survey of 188 IT decision-makers. The proposed Open Source Risk Adoption Model offers novel insights on the importance of the perceived risk antecedents. Our research brings new theoretical contributions, such as understanding the perceived IT security risk (PISR) relationship with adoption intention (AI) in the OSS context, for researchers and important insights for IT information professionals. We have found that IT security risk has a significant role in OSS adoption intention. Our results offer possible future research directions and extend existing theoretical understanding of OSS adoption.

This study aimed to conduct operation performance evaluations of Taiwan’s national hospitals during the period 2005–2008 and also propose appropriate suggestions for operation performance improvements. 28 national hospitals were selected as study objects and six input variables and three output variables were filtered as evaluation measures. Data envelopment analysis model was used for annual operation performance evaluation, and Malmquist productivity index for intertemporal operation performance change analysis. In addition, a performance scatter diagram and a strategy management matrix were utilized to synthetically analyze all kinds of operation performance data and accordingly improvement suggestions were proposed. Several findings were explored, including (1) nearly 60% of national hospitals ran an inefficient performance; (2) a significant gap between urban and nonurban hospitals did exist in health care resources allocation and medical service outputs, which reflected the negative public opinion about regional medical care resources gap in Taiwan; (3) other evidence depicted the expectation that the government of Taiwan has already wasted a lot of medical resources in the operation of Taiwan’s national hospitals; (4) 70–80% of Taiwan’s national hospitals ran a constant returns-of-scale operation, while the remainder ran a decreasing returns-of-scale operation; and (5) a extreme disparity existed in the total productivity, which rationally conjectured the cause came from the bad operation of several individual national hospitals. At last, two main suggestions advocated: (a) the government of Taiwan should reconsider the medical care resources allocation between urban and nonurban hospitals; and (b) hospital operation performance should be regarded as one of the main prerequisites for government budget subsidies so as to stimulate operation performance improvement and self-sufficiency operation.

The research on blog emotion analysis and recognition has become increasingly important in recent years. In this study, based on the Chinese blog emotion corpus (Ren-CECps), we analyze and compare blog emotion visualization from different text levels: word, sentence, and paragraph. Then, a blog emotion visualization system is designed for practical applications. Machine learning methods are applied for the implementation of blog emotion recognition at different textual levels. Based on the emotion recognition engine, the blog emotion visualization interface is designed to provide a more intuitive display of emotions in blogs, which can detect emotion for bloggers, and capture emotional change rapidly. In addition, we evaluated the performance of sentence emotion recognition by comparing five classification algorithms under different schemas, which demonstrates the effectiveness of the Complement Naive Bayes model for sentence emotion recognition. The system can recognize multi-label emotions in blogs, which provides a richer and more detailed emotion expression.