International Journal of Information Technology & Decision Making

In the field of artificial intelligence, human–computer interaction (HCI) technology and its related intelligent robot technologies are essential and interesting contents of research. From the perspective of software algorithm and hardware system, these above-mentioned technologies study and try to build a natural HCI environment. The purpose of this research is to provide an overview of HCI and intelligent robots. This research highlights the existing technologies of listening, speaking, reading, writing, and other senses, which are widely used in human interaction. Based on these same technologies, this research introduces some intelligent robot systems and platforms. This paper also forecasts some vital challenges of researching HCI and intelligent robots. The authors hope that this work will help researchers in the field to acquire the necessary information and technologies to further conduct more advanced research.

Clustering is a complex unsupervised method used to group most similar observations of a given dataset within the same cluster. To guarantee high efficiency, the clustering process should ensure high accuracy and low complexity. Many clustering methods were developed in various fields depending on the type of application and the data type considered. Categorical clustering considers segmenting a dataset in which the data are categorical and were widely used in many real-world applications. Thus several methods were developed including hard, fuzzy and rough set-based methods. In this survey, more than 30 categorical clustering algorithms were investigated. These methods were classified into hierarchical and partitional clustering methods and classified in terms of their accuracy, precision and recall to identify the most prominent ones. Experimental results show that rough set-based clustering methods provided better efficiency than hard and fuzzy methods. Besides, methods based on the initialization of the centroids also provided good results.

By 2018, business analytics (BA), believed by global CIOs to be of strategic importance, had for years been their top priority. It is also a focus of academic research, as shown by a large number of papers, books, and research reports. On the other hand, the BA domain suffers from several incorrect, imprecise, and incomplete notions. New areas and concepts emerge quickly; making it difficult to ascertain their structure. BA-related taxonomies play a crucial role in analyzing, classifying, and understanding related objects. However, according to the literature on taxonomy development in information systems (IS), in most cases the process is ad hoc. BA taxonomies and frameworks are available in the literature; however, some are excessively general frameworks with a high-level conceptual focus, while others are application or domain-specific. Our paper aims to present a novel semi-automatic method for taxonomy development and maintenance in the field of BA using content analysis and text mining. The contribution of our research is threefold: (1) the taxonomy development method, (2) the draft taxonomy for BA, and (3) identifying the latest research areas and trends in BA.

To improve the quality of service and network performance for the Flash P2P video-on-demand, the prediction Flash P2P network traffic flow is beneficial for the control of the network video traffic. In this paper, a novel prediction algorithm to forecast the traffic rate of Flash P2P video is proposed. This algorithm is based on the combination of the ensemble local mean decomposition (ELMD) and the generalized autoregressive conditional heteroscedasticity (GARCH). The ELMD is used to decompose the original long-related flow into the summation of the short-related flow. Then, the GRACH is utilized to predict the short-related flow. The developed algorithm is tested in a university’s campus network. The predicted results show that our proposed method can further achieve higher accuracy than those obtained by existing algorithms, such as GARCH and Local Mean Decomposition and Generalized AutoRegressive Conditional Heteroskedasticity (LMD-GARCH) while keeping lower computational complexity.

The linguistic $D$ numbers (LDNs) can express the fuzzy evaluation information more easily and precisely by combining the advantages of linguistic terms (LTs) and $D$ numbers (DNs). Existing researches on the fuzzy entropy of LDNs are rare, and most of the definitions of fuzzy entropy for LDNs are unreasonable. In view of this research gap, this paper improves the definition of fuzzy entropy of LDNs, which simultaneously considers the effects of confidence degrees and LTs on the value of fuzzy entropy in LDNs. Then, the weights of attributes can be calculated by the improved fuzzy entropy. Further, a new combination rule for LDNs is also presented in this paper. Based on these studies, we extend the traditional Vlsekriterijumska Optimizacija I Kompromisno Resenje (VIKOR) method to the LDNs and develop the LD-VIKOR method. The proposed LD-VIKOR method is convenient to handle MADM problems with unknown attributes weights under the environment of LDNs. Finally, we verify the validity and reliability of the proposed method by an illustrative example, and analyze the advantages of the proposed method by comparing it with other existing MADM methods.

Since 2013, China has become the world’s largest gold producer and consumer. To gain the corresponding global pricing power in gold, many actions have been taken by China in recent years, including the International Board at Shanghai Gold Exchange, Shanghai-Hong Kong Gold Connect and Shanghai Gold Fix. Our work studies the dependence structure between China’s and international gold price and examines whether these moves are changing the dependence structure. We use GARCH-copula models to detect the dynamic dependence and tail dependence. The research period is set to contain the Financial Crisis in 2008, the dramatical plunge of gold price in 2013 and a series of black swan events in 2016. The empirical study shows that some event driven dependence structure breaks are statistically insignificant. And the time-varying Symmetrized Joe-Clayton copula is the best copula to model the dependence structure based on AIC value. Finally, an example of applications of this dependence structure is given by estimating the VaR of an equally weighted portfolio with a simulation-based method.

In time series modeling, one problem is to identify a small number of influential factors to explain variations in the variable of interest. With a vast number of possible factors available, suitable features need to be identified to yield multi-factor models with good explanatory power. In this paper, we propose a novel subset selection method which makes use of the properties in the frequency domain environment. The proposed system ensures key patterns in the target variable be sought and suitable factors be selected based on frequency peaks in common. It can perform well even when the number of factors is significantly greater than the sample size. Moreover, a very important feature of the proposed system is the capability of handling factors with different timeframes, which is lacking in existing methods. We demonstrate the system via several examples with dataset from finance, economic, road traffic and air pollution.

This paper introduces a novel approach to support decision-making by combining the Weighted Influence Nonlinear Gauge System (WINGS) method with interval arithmetic. This approach allows to include uncertain judgments and/or different opinions in a decision process. Our research aims at increasing the ability of WINGS to model decisions in situations of uncertainty and at extending the reach of its practical applications. The new, relatively simple and transparent method can become a useful and practical tool for the decision makers. Mathematical correctness of the proposed methodology is proven.

Based on the new method, a procedure for solving a complex decision problem is created. Its applicability is illustrated by two case studies. Choosing the best option for the organization’s competitive position in a health-care organization shows how the proposed method works with uncertain judgments. Its usefulness for group decision-making is illustrated by applying it to a decision concerning allocation of public funds for sport development in a small commune.

The theoretical and practical researches on Visual Analytics for intelligent decision-making tasks have remarkably advanced in the past few years. Intelligent Decision Support Systems (IDSS) introduce effective and efficient paths from raw data to decision by involving visualization and data mining technologies. Data mining-based DSS produces potentially interesting patterns from data. The transition from extracted patterns to knowledge is a delicate task. In this context, we propose to adapt a common visual analytics process for creating a path that enables the user (decision-maker) to automatically explore and visually extract insights by interacting with the patterns. This proposal is inspired from integrating traditional visual analytics concepts with the mental model of knowledge visualization. The idea is to combine an automatic and visual analysis of patterns to generate knowledge for the purpose of decision-making. To validate our proposal, we have applied it to a medical case study for the fight against Nosocomial Infections in Intensive Care Units. The developed platform was evaluated according to the utility and usability dimensions.

This paper proposes a new approach to improve the performance of Learning Algorithm for Multivariable Data Analysis (LAMDA). This algorithm can be used for supervised and unsupervised learning, based on the calculation of the Global Adequacy Degree (GAD) of one individual to a class, through the contributions of all its descriptors. LAMDA has the capability of creating new classes after the training stage. If an individual does not have enough similarity to the preexisting classes, it is evaluated with respect to a threshold called the Non-Informative Class (NIC), this being the novelty of the algorithm. However, LAMDA has problems making good classifications, either because the NIC is constant for all classes, or because the GAD calculation is unreliable. In this work, its efficiency is improved by two strategies, the first one, by the calculation of adaptable NICs for each class, which prevents that correctly classified individuals create new classes; and the second one, by computing the Higher Adequacy Degree (HAD), which grants more robustness to the algorithm. LAMDA-HAD is validated by applying it in different benchmarks and comparing it with LAMDA and other classifiers, through a statistical analysis to determinate the cases in which our algorithm presents a better performance.

The Amazon region has the largest hydrographic basin in the world and a favorable geographical location as a strategic export point but does not satisfactorily exploit its potential for transport and cargo flow, mainly as a differential factor of logistic competitiveness. For the waterway mode, cargo transportation is predominantly carried out by pushed convoys. The objective of this research is to evaluate and select, the best river train for the outflow of iron ore on the Marabá – Vila do Conde stretch, through a multiple criteria decision-making approach. To meet the objectives, a methodological framework is proposed using a hybrid AHP-DEMATEL method, based on a statistical analysis and to guarantee the reliability of the results for the proposed alternatives and not only considering the preference function, the ELECTRE method was used and also a parametric comparison of the AHP results. The main goals of this work were achieved and it was possible to infer that the combination of these methods has, in addition to the evident efficacy, a high reliability in the water transport area, being able to be used with accuracy of diverse forms in the academic area, as well as in industrial applications. Therefore, it can be affirmed that the best-pushed convoy configuration for the iron ore runoff in the Marabá – Vila do Conde stretch is composed by six barges in the arrangement of two columns and three rows with the measurements, by barge, with length of 60.96m; beam of 13.75m; and a 5m molded depth.

The single-valued neutrosophic set (SVNS) is considered as an attractive tool for handling highly uncertain and vague information. With this regard, different from the most current distance-based technique for order preference by similarity to ideal solution (TOPSIS) methods, this study proposes a correlation-based TOPSIS model for addressing the single-valued neutrosophic (SVN) multiple attribute decision making (MADM) problems. To achieve this aim, we first develop a novel conception of SVN correlation coefficient, whose significant feature is that it lies in the interval [$-1$,1], which is in accordance with the classical correlation coefficient in statistics, whereas all the existing SVN correlation coefficients in the literature are within unit interval [0,1]. Afterwards, a weighted SVN correlation coefficient is also introduced to infuse the importance of attributes. Moreover, a correlation-based comprehensive index is further proposed to establish the central structure of TOPSIS model, called the SVN correlation-based TOPSIS approach. Finally, a numerical example and relevant comparative analysis are implemented to explain the applicability and effectiveness of the mentioned methodology.