International Journal of Information Technology & Decision Making

In the past, there has been significant research on tracking objects based on features that characterize objects. However, once the features to detect the object are not available from the position of a tracker, it is not clear how to track the object. Our goal in this paper is to track an object even though the object leaves the field-of-view. In this paper, we first describe the problems related to this type of tracking. We then explain our approach that has two phases: Real object tracking and virtual object tracking. We mostly focus on virtual object tracking that requires reorientation, distance estimation, traveling the computed distance, estimating the direction for turn, and turning into the direction. We show our results by building a robot that achieves these.

Collaborative filtering (CF) is one of the most prevalent recommendation techniques, providing personalized recommendations to users based on their previously expressed preferences and those of other similar users. Although CF has been widely applied in various applications, its applicability is restricted due to the data sparsity, the data inadequateness of new users and new items (cold start problem), and the growth of both the number of users and items in the database (scalability problem). In this paper, we propose an efficient iterative clustered prediction technique to transform user-item sparse matrix to a dense one and overcome the scalability problem. In this technique, spectral clustering algorithm is utilized to optimize the neighborhood selection and group the data into users' and items' clusters. Then, both clustered user-based and clustered item-based approaches are aggregated to efficiently predict the unknown ratings. Our experiments on MovieLens and book-crossing data sets indicate substantial and consistent improvements in recommendations accuracy compared to the hybrid user-based and item-based approach without clustering, hybrid approach with k-means and singular value decomposition (SVD)-based CF. Furthermore, we demonstrated the effectiveness of the proposed iterative technique and proved its performance through a varying number of iterations.

This paper, using goal programming (GP) and the analytic hierarchy process (AHP), proposes an integrated methodology to aid decision makers in (1) evaluating, screening and selecting best suppliers from among an exhaustive list of available suppliers and (2) determining the amount to be purchased from the selected suppliers. Along the supply chain, the suppliers, by being situated at the upstream of the chain, play a crucial role in successful management of the entire (subsequent) members of the chain and can have a significant impact on the efficiency and effectiveness of the activities of the rest of the chain, and ultimately, on the delivery of the desired products/services. Thus supplier selection can certainly contribute greatly to a firm's competitive advantage and its organizational success. This study suggests a screening and evaluation method, named supplier priority index matrix, to eliminate the least qualified suppliers and to choose the most promising ones. The remaining potential suppliers are evaluated and the best are selected from among them. Finally the AHP and GP are used to allocate the firm's total supplies among the chosen/selected suppliers.

This study develops a diagnosing model to examine the outcomes of assets write-off in enriching the literatures of assets impairment. Prior studies employed the Logit, linear and Tobit regression models to classify the determination of assets impairment and to diagnose the magnitude of the impairment, respectively. However, the drivers of assets write-off are somewhat complicated explicitly or implicitly, these models are unlikely to provide fairly satisfactory results. To improve the diagnosis, the Random Forests model is used for the classification determining and the magnitude diagnosing of assets impairment in this study. The result reveals that the Random Forests model outperforms the Logit and linear regression models in each case with variables selected by individual wrapping approach. This study also demonstrates diagnostic checks for both models with similar selected variables. The results are robust to these various specifications.

This paper focuses on the problem of resource allocation through data envelopment analysis. We propose three integrated models for allocating resources. The first model aims at minimizing the input consumption, the second one aims at maximizing the total outputs within the current resources, and the last one aims at maximizing the total outputs using the predicted resources in the next production season. Since the number of inputs or outputs is usually more than one, the abovementioned issue is often a multiple objective linear programming (MOLP) problem. Through the proportion of inputs (outputs) of new decision making unit (DMU) to the total inputs (outputs) of all old DMUs, we transform the MOLP problem into a single objective linear programming model. We assume that decision maker must ensure that the expected outputs of each DMU after allocation in the next production season are not less than this production season. All these proposed models have the same advantage that the results gained from the models are Pareto efficient. A numerical example of 25 supermarkets is used to illustrate our proposed approach.

Generality of the model which is used in Dezert–Smarandache Theory (DSmT) rather than other fusion algorithms such as Dempster–Shafer theory and capability of DSmT for dealing with highly conflict problems are two main reasons to prefer DSmT for decision-making systems. Aggregated uncertainty measure, which is called AU measure, has been introduced for Dempster–Shafer theory as one of the best presented ways to quantify the total uncertainty or the ambiguity of a belief function. Since AU cannot be applied to DSmT, two generalized aggregated uncertainty measures for DSmT, which are called GAU1 measure and GAU2 measure, are proposed in this paper. The GAU1 measure is developed by extension of the frame of discernment with distinct sub-events. The GAU2 measure is developed by considering new conditions on probability assignments which are used in the uncertainty measure. The new conditions are the difference of this measure with AU measure. A rigorous discussion is presented to validate that the proposed uncertainty measures holds on the requirements for an uncertainty measure. Finally evaluation of uncertainty in a DSmT-based localization problem is presented to show how to apply the generalized uncertainty measures, GAU1 and GAU2.

This research analyzes the influence of dispersion of valuations on financial markets, taking several aspects of real financial market into consideration (such as financial constraints, investment strategies and so on). As a result of intensive experiments in the market, we made the following findings: (1) Dispersion of fundamentalists' valuations has little effect on the market when financial constraints are absent; (2) When financial constraints — such as short-sale constraints — are introduced, certain situations arise in which deviations from fundamental values become larger, according to the level of the dispersion of valuations; (3) A passive investment strategy, as is consistent with traditional financial theory, is valid even when the introduction of financial constraints causes market prices to deviate significantly from fundamental values. These results contribute to clarifying the mechanism of price fluctuations in financial markets and are notable from both academic and practical view points.

Fuzzy time series has been applied to forecast various domain problems because of its capability to deal with vagueness and incompleteness inherent in data. However, most existing fuzzy time series models cannot cope with multi-attribute time series and remain too subjective in the partition of the universe of discourse. Moreover, these models do not consider the trend factor and the corresponding external time series, which are highly relevant to target series. In the current paper, a heuristic bivariate model is proposed to improve forecasting accuracy, and the proposed model applies fuzzy c-means clustering algorithm to process multi-attribute fuzzy time series and to partition the universe of discourse. Meanwhile, the trend predictors are extracted in the training phase and utilized to select the order of fuzzy relations in the testing phase. Finally, the proper full use of the external series to assist forecasting is discussed. The performance of the proposed model is tested using actual time series including the enrollments at the University of Alabama, the Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX) and a sensor dataset. The experimental results show that the proposed model can be utilized for multi-attribute time series and significantly improves the average MAER to 1.19% when compared with other forecasting models.

Classification algorithm selection is an important issue in many disciplines. Since it normally involves more than one criterion, the task of algorithm selection can be modeled as multiple criteria decision making (MCDM) problems. Different MCDM methods evaluate classifiers from different aspects and thus they may produce divergent rankings of classifiers. The goal of this paper is to propose an approach to resolve disagreements among MCDM methods based on Spearman's rank correlation coefficient. Five MCDM methods are examined using 17 classification algorithms and 10 performance criteria over 11 public-domain binary classification datasets in the experimental study. The rankings of classifiers are quite different at first. After applying the proposed approach, the differences among MCDM rankings are largely reduced. The experimental results prove that the proposed approach can resolve conflicting MCDM rankings and reach an agreement among different MCDM methods.