Information, Volume 11, Issue 3 (March 2020) – 49 articles

Industry 4.0 has managed to attract lots of attention from researchers because of the benefits it has to offer and many studies about Industry 4.0, as well as readiness and maturity models, are available from the technological point of view. Meanwhile, the organizational culture dimension of Industry 4.0 has received little to no attention. The aim of this paper is to find out whether or not the size and type of an organization influence the innovative culture and consequently the readiness of the organization for implementing industry 4.0. Results show that the innovative organizational culture according to the index of organizational culture does not depend on the size of an organization but to some degree depends on the type of the organization. Full article

Research on external human–machine interfaces (eHMIs) has recently become a major area of interest in the field of human factors research on automated driving. The broad variety of methodological approaches renders the current state of research inconclusive and comparisons between interface designs impossible. To date, there are no standardized test procedures to evaluate and compare different design variants of eHMIs with each other and with interactions without eHMIs. This article presents a standardized test procedure that enables the effective usability evaluation of eHMI design solutions. First, the test procedure provides a methodological approach to deduce relevant use cases for the evaluation of an eHMI. In addition, we define specific usability requirements that must be fulfilled by an eHMI to be effective, efficient, and satisfying. To prove whether an eHMI meets the defined requirements, we have developed a test protocol for the empirical evaluation of an eHMI with a participant study. The article elucidates underlying considerations and details of the test protocol that serves as framework to measure the behavior and subjective evaluations of non-automated road users when interacting with automated vehicles in an experimental setting. The standardized test procedure provides a useful framework for researchers and practitioners. Full article

To improve utilization of text storage resources and efficiency of data transmission, we proposed two syllable-based Uyghur text compression coding schemes. First, according to the statistics of syllable coverage of the corpus text, we constructed a 12-bit and 16-bit syllable code tables and added commonly used symbols—such as punctuation marks and ASCII characters—to the code tables. To enable the coding scheme to process Uyghur texts mixed with other language symbols, we introduced a flag code in the compression process to distinguish the Unicode encodings that were not in the code table. The experiments showed that the 12-bit coding scheme had an average compression ratio of 0.3 on Uyghur text less than 4 KB in size and that the 16-bit coding scheme had an average compression ratio of 0.5 on text less than 2 KB in size. Our compression schemes outperformed GZip, BZip2, and the LZW algorithm on short text and could be effectively applied to the compression of Uyghur short text for storage and applications. Full article

Inferring customers’ preferences and recommending suitable products is a challenging task for companies, although recommender systems are constantly evolving. Loyalty is an indicator that measures the preference relationship between customers and products in the field of marketing. To this end, the aim of this study is to explore whether customer loyalty can improve the accuracy of the recommender system. Two algorithms based on complex networks are proposed: a recommendation algorithm based on bipartite graph and PersonalRank (BGPR), and a recommendation algorithm based on single vertex set network and DeepWalk (SVDW). In both algorithms, loyalty is taken as an attribute of the customer, and the relationship between customers and products is abstracted into the network topology. During the random walk among nodes in the network, product recommendations for customers are completed. Taking a real estate group in Malaysia as an example, the experimental results verify that customer loyalty can indeed improve the accuracy of the recommender system. We can also conclude that companies are more effective at recommending customers with moderate loyalty levels. Full article

Digital and networking technologies are increasingly used to predict who is at risk of attempting suicide. Such digitalized suicide prevention within and beyond mental health care raises ethical, social and legal issues for a range of actors involved. Here, I will draw on key literature to explore what issues (might) arise in relation to digitalized suicide prevention practices. I will start by reviewing some of the initiatives that are already implemented, and address some of the issues associated with these and with potential future initiatives. Rather than addressing the breadth of issues, however, I will then zoom in on two key issues: first, the duty of care and the duty to report, and how these two legal and professional standards may change within and through digitalized suicide prevention; and secondly a more philosophical exploration of how digitalized suicide prevention may alter human subjectivity. To end with the by now famous adagio, digitalized suicide prevention is neither good nor bad, nor is it neutral, and I will argue that we need sustained academic and social conversation about who can and should be involved in digitalized suicide prevention practices and, indeed, in what ways it can and should (not) happen. Full article

In this study, a total of 94,511 surveys were used for the analysis, using raw data from the recent 3 years (2016–2018) of the Korea Youth Risk Behavior Survey (KYRBS) to compare differences in obesity rates based on eating behavior and physical activity among high school students. The cross-analysis and logistic regression were performed in the composite sample design using the SPSS 25.0 statistical package. The results are as follows. First, the more physical activity male students have, the lower the obesity rate. Female students, however, showed a minor difference. Second, the group that drinks the soda was 1158 times more likely to be obese than the group that does not drink the soda, and third, the group that eats the fast food was 1129 times more likely to be obese than the group that does not. Fourth, students belonging to male or female schools had a 1230 times higher obesity rate than coeducational students. Fifth, the obesity rate was 1150 times higher for second grade and 1263 times higher for third grade compared to the first grade. Finally, there was no significant difference related to the nutrition education. It is suggested that, to fight youth obesity in high school students, institutions need to raise public awareness of this problem through information campaigns aimed at improving and fostering potentially existing educational measures. Full article

Early detection of the security incidents and correct forecasting of the attack development is the basis for the efficient and timely response to cyber threats. The development of the attack depends on future steps available to the attackers, their goals, and their motivation—that is, the attacker “profile” that defines the malefactor behaviour in the system. Usually, the “attacker profile” is a set of attacker’s attributes—both inner such as motives and skills, and external such as existing financial support and tools used. The definition of the attacker’s profile allows determining the type of the malefactor and the complexity of the countermeasures, and may significantly simplify the attacker attribution process when investigating security incidents. The goal of the paper is to analyze existing techniques of the attacker’s behaviour, the attacker’ profile specifications, and their application for the forecasting of the attack future steps. The implemented analysis allowed outlining the main advantages and limitations of the approaches to attack forecasting and attacker’s profile constructing, existing challenges, and prospects in the area. The approach for attack forecasting implementation is suggested that specifies further research steps and is the basis for the development of an attacker behaviour forecasting technique. Full article

Together with the disruptive development of modern sub-symbolic approaches to artificial intelligence (AI), symbolic approaches to classical AI are re-gaining momentum, as more and more researchers exploit their potential to make AI more comprehensible, explainable, and therefore trustworthy. Since logic-based approaches lay at the core of symbolic AI, summarizing their state of the art is of paramount importance now more than ever, in order to identify trends, benefits, key features, gaps, and limitations of the techniques proposed so far, as well as to identify promising research perspectives. Along this line, this paper provides an overview of logic-based approaches and technologies by sketching their evolution and pointing out their main application areas. Future perspectives for exploitation of logic-based technologies are discussed as well, in order to identify those research fields that deserve more attention, considering the areas that already exploit logic-based approaches as well as those that are more likely to adopt logic-based approaches in the future. Full article

Privacy preserving data publishing has received considerable attention for publishing useful information while preserving data privacy. The existing privacy preserving data publishing methods for multiple sensitive attributes do not consider the situation that different values of a sensitive attribute may have different sensitivity requirements. To solve this problem, we defined three security levels for different sensitive attribute values that have different sensitivity requirements, and given an $L_{sl}$ -diversity model for multiple sensitive attributes. Following this, we proposed three specific greed algorithms based on the maximal-bucket first (MBF), maximal single-dimension-capacity first (MSDCF) and maximal multi-dimension-capacity first (MMDCF) algorithms and the maximal security-level first (MSLF) greed policy, named as MBF based on MSLF (MBF-MSLF), MSDCF based on MSLF (MSDCF-MSLF) and MMDCF based on MSLF (MMDCF-MSLF), to implement the $L_{sl}$ -diversity model for multiple sensitive attributes. The experimental results show that the three algorithms can greatly reduce the information loss of the published microdata, but their runtime is only a small increase, and their information loss tends to be stable with the increasing of data volume. And they can solve the problem that the information loss of MBF, MSDCF and MMDCF increases greatly with the increasing of sensitive attribute number. Full article

Compared with the conventional simultaneous wireless information and power transfer (SWIPT) based relaying with “harvest-then-forward” protocol, the battery-assisted SWIPT relaying is more practical and powerful due to the joint use of the harvested energy and supplementary battery. However, to the best of our knowledge, the performance of a battery-assisted power splitting (PS)-SWIPT decode-and-forward (DF) relay system has not been studied. In this paper, for a given amount of energy from the relay’s battery, we propose to maximize the outage and ergodic capacities by optimizing the static and dynamic PS ratios that rely on statistical and instantaneous channel state information (CSI), respectively, and derive their corresponding outage and ergodic capacities. Computer simulations validate our analytical results and demonstrate the advantages of the dynamic PS over the static PS in terms of the outage and ergodic capacities, as well as the energy efficiency. Full article

The article outlines the main concept and examples of mathematical models needed to estimate system potential and digitalization performance indicators. Such an estimation differs in that it is performed with predictive mathematical models. The purpose of such an estimation is to enable a set of problems of system design and functional design of information technologies to be solved as mathematical problems, predictively and analytically. The hypothesis of the research is that the quality of system functioning in changing conditions can be evaluated analytically, based on predictive mathematical models. We suggested a property of the system potential (or system capability) that describes the effects of the compliance of changing system functioning with changing conditions analytically and predictively. Thus, it describes the performance of the use of information operations to realize functioning in changing conditions. The example includes system’s environment graph-theoretic models and system’s models regarding IT use to react to changing environments. As a result of the suggested models and methods, the quantitative estimation of system potential regarding information technology use becomes possible depending on the parameters and variables of the problems to be solved. Use cases of problems decision with the use of such indicators include choosing optimal information technology implementation and the synthesis of information operation characteristics. Full article

The main drivers for the continuous development of Vehicular ad-hoc Networks (VANETs) are safety applications and services. However, in recent years, new interests have emerged regarding the introduction of new applications and services for non-urgent content (e.g., videos, ads, sensing and touristic information) dissemination. However, there is a lack of real studies considering content dissemination strategies to understand when and to whom the content should be disseminated using real vehicular traces gathered from real vehicular networks. This work presents a realistic study of strategies for dissemination of non-urgent content with the main goal of improving content delivery as well as minimizing network congestion and resource usage. First, we perform an exhaustive network characterization. Then, several content strategies are specified and evaluated in different scenarios (city center and parking lot). All the obtained results show that there are two content distribution strategies that clearly set themselves apart due to their superior performance: Local Rarest Bundle First and Local Rarest Generation First. Full article

In this paper we examine the recent situation of advocacy through game accessibility of video games for people with physical disabilities and its challenges in Japan. We conducted an email survey among the dominant gaming companies in Japan. We discovered that most gaming companies developing games for those with disabilities have actually attempted to provide some consideration to people with disabilities through setting additional special functions such as “easy mode” and “support for visual disabilities”. One out of five companies could be categorized as having successfully developed games based on the concept of specialized assistive technology, and the remaining companies could be categorized as having developed games for disabilities based on the concept of barrier-reducing goods. We also found out that the major Japanese gaming corporations had a tendency not to be active in communicating with the disability communities. Our results might indicate that stronger interaction with the gaming companies is one direct way to enhance further development of games for people with disabilities. In addition to that, considering that games are a significant form of social participation for people with disabilities, more social understanding is required in the Japanese domestic context. Full article

Nowadays, it is possible to buy clothing using online platforms, either by accessing online brand stores, general online stores or circular economy platforms. This paper presents a study on mobile applications that support online commerce for clothing, focusing on the review of the mobile applications with features that are characteristic of the circular economy paradigm. Findings include the fact that almost all the mobile applications analysed have pictures illustrative of the clothes and accessories that are available for trading as well as their brief description. Furthermore, this paper presents a study of various scientific articles about the circular economy of clothes and how it can be beneficial to the future of the environment. It is a junction with a Web platform for its growth and its disclosure. The paper builds conclusions upon the assumption that the circular economy is a growing business that is part of a sustainable development where the main goal is to reduce the environmental impact. The paper proposes the analysis of an innovative taxonomy of mobile applications about the circular economy. Full article

This paper focuses on business financial health evaluation with the use of selected mathematical and statistical methods. The issue of financial health assessment and prediction of business failure is a widely discussed topic across various industries in Slovakia and abroad. The aim of this paper was to formulate a data envelopment analysis (DEA) model and to verify the estimation accuracy of this model in comparison with the logit model. The research was carried out on a sample of companies operating in the field of heat supply in Slovakia. For this sample of businesses, we selected appropriate financial indicators as determinants of bankruptcy. The indicators were selected using related empirical studies, a univariate logit model, and a correlation matrix. In this paper, we applied two main models: the BCC DEA model, processed in DEAFrontier software; and the logit model, processed in Statistica software. We compared the estimation accuracy of the constructed models using error type I and error type II. The main conclusion of the paper is that the DEA method is a suitable alternative in assessing the financial health of businesses from the analyzed sample. In contrast to the logit model, the results of this method are independent of any assumptions. Full article

After a literature review published by Nascimento et. al. (2017), the research team noticed the lack of studies focused on game controllers’ accessibility during use by children with Down syndrome. In view of that, this research describes a mobile game development and its usability analyses, which were created to evaluate the accessibility of touchscreen gestural interfaces. The methodology was organized into three steps: bibliographic research and the definition of the project guidelines, the game development, and its evaluation. The guidelines used were based on a study made by Nascimento et. al. (2019) of the impairments that children can have, their game preferences found on Prena’s article (2014), games accessibility guidelines for people with intellectual deficiency from the Includification Book (2012), a manual of touchscreen gestural interfaces from Android and iOS and a game development framework from Schuytema (2008). Then, for the usability analyses, the team decided to first submit the game to a group of experts in order to make some improvements before submitting it to the audience. In this way, two evaluations were done, a heuristic test with usability specialists and a cognitive walkthrough with health professionals. The list of heuristics used on the tests was created by a mash up of the Breyer evaluation (2008) and the recommendations of the Able Games Association (2012) and the cognitive one followed the Preece, Sharp and Rogers (2007) recommendations. The results found reveal some challenges in the field and adjustments, mainly in the narrative, game goals and interface feedback, that should be addressed as soon as possible. Full article

Predicting the categories of actions in partially observed videos is a challenging task in the computer vision field. The temporal progress of an ongoing action is of great importance for action prediction, since actions can present different characteristics at different temporal stages. To this end, we propose a novel multi-task deep forest framework, which treats temporal progress analysis as a relevant task to action prediction and takes advantage of observation ratio labels of incomplete videos during training. The proposed multi-task deep forest is a cascade structure of random forests and multi-task random forests. Unlike the traditional single-task random forests, multi-task random forests are built upon incomplete training videos annotated with action labels as well as temporal progress labels. Meanwhile, incorporating both random forests and multi-task random forests can increase the diversity of classifiers and improve the discriminative power of the multi-task deep forest. Experiments on the UT-Interaction and the BIT-Interaction datasets demonstrate the effectiveness of the proposed multi-task deep forest. Full article

Most people are using their smartphones daily and carry them all the time. Therefore, mobile learning applications can be integrated in daily routines to make learning a part of daily life. While numerous mobile learning applications exist, each with their own goal, our aim was to explore the possibility of creating an engaging mobile environment that could be useful for informal learning, as well as for other forms of learning, i.e., non-formal and formal learning. The result is TICKLE, a playful learning environment for youngsters. It is a mobile location-based smartphone application that offers youngsters an interactive environment for exploring their surroundings. It offers cards related to physical locations, which can be collected by performing small challenges (based on the principles of micro learning). A design science research approach has been used to develop this software environment. Persuasive techniques and gamification are used to stimulate usage. Furthermore, a personalized approach is applied. The environment was evaluated by means of formative evaluations in different contexts. We obtained positive results and received useful feedback to improve and extend the application. We can conclude that, in the context of these evaluations, the app was usable for youngsters and able to engage them, and we see indications that it may be able to increase the intrinsic motivation and learning capacity of youngsters. In addition, our demonstrations show that the app is usable in different contexts and for different purposes. In this way, the environment can be used to offer youngsters appealing learning related experiences. Full article

Competition is a basic element of our society. It drives us to rise above previously perceived limitations, increases our engagement and makes the world more interesting. Competition rewards our existing skills and prompts us to identify and improve our weaker skills. In games, player engagement is achieved, at least in part, by providing him/her with competition at the right amount of difficulty. Achieving and maintaining this exact level of challenge is one of the most difficult tasks for a game designer. The use of Dynamic Difficulty Adjustment techniques allows the game to dynamically adjust the challenge according to player performance, therefore keeping him/her always on edge, immersed and fully active. New information can then be more easily acquired, which is especially important in Serious Games. This paper describes how DDA techniques were used to create two strategic, goal-oriented computer-controlled (CC) players in order to deliver a higher level of competitiveness for the user in [email protected], a Serious Game aimed at developing entrepreneurship skills. As a result, the strength of the computer controlled player increased by more than 100%. By developing a good strategy for the AI and using DDA the game includes now a powerful opponent which has increased the engagement level of the player. Full article

Accurate medical images analysis plays a vital role for several clinical applications. Nevertheless, the immense and complex data volume to be processed make difficult the design of effective algorithms. The first aim of this paper is to examine this area of research and to provide some relevant reference sources related to the context of medical image analysis. Then, an effective hybrid solution to further improve the expected results is proposed here. It allows to consider the benefits of the cooperation of different complementary approaches such as statistical-based, variational-based and atlas-based techniques and to reduce their drawbacks. In particular, a pipeline framework that involves different steps such as a preprocessing step, a classification step and a refinement step with variational-based method is developed to identify accurately pathological regions in biomedical images. The preprocessing step has the role to remove noise and improve the quality of the images. Then the classification is based on both symmetry axis detection step and non linear learning with SVM algorithm. Finally, a level set-based model is performed to refine the boundary detection of the region of interest. In this work we will show that an accurate initialization step could enhance final performances. Some obtained results are exposed which are related to the challenging application of brain tumor segmentation. Full article

Criminals use online social networks for various activities by including communication, planning, and execution of criminal acts. They often employ ciphered posts using slang expressions, which are restricted to specific groups. Although literature shows advances in analysis of posts in natural language messages, such as hate discourses, threats, and more notably in the sentiment analysis; research enabling intention analysis of posts using slang expressions is still underexplored. We propose a framework and construct software prototypes for the selection of social network posts with criminal slang expressions and automatic classification of these posts according to illocutionary classes. The developed framework explores computational ontologies and machine learning (ML) techniques. Our defined Ontology of Criminal Expressions represents crime concepts in a formal and flexible model, and associates them with criminal slang expressions. This ontology is used for selecting suspicious posts and decipher them. In our solution, the criminal intention in written posts is automatically classified relying on learned models from existing posts. This work carries out a case study to evaluate the framework with 8,835,290 tweets. The obtained results show its viability by demonstrating the benefits in deciphering posts and the effectiveness of detecting user’s intention in written criminal posts based on ML. Full article

Teaching software engineering in its many different forms using traditional teaching methods is difficult. Serious games can help overcome these challenges because they allow real situations to be simulated. However, the development of serious games is not easy and, although there are good practices for relating game design patterns to teaching techniques, there is no methodology to support its use in a specific context such as software engineering. This article presents a case study to validate a methodology that links the Learning and Teaching Functions (LTF) to the Game Design Patterns (PIB) in the context of Software Engineering Education. A serious game was developed from scratch using this methodology to teach software estimation (a specific topic of software engineering). An experiment was carried out to validate the effectiveness of the game by comparing the results of two different groups of students. The results indicate that the methodology can help to develop effective educational games on specific learning topics. Full article

Mobile applications (apps) on IOS and Android devices are mostly maintained and updated via Apple Appstore and Google Play, respectively, where the users are allowed to provide reviews regarding their satisfaction towards particular apps. Despite the importance of user reviews towards mobile app maintenance and evolution, it is time-consuming and ineffective to dissect each individual negative review. In addition, due to the different app update strategies, it is uncertain that each update can be accepted well by the users. This study aims to provide an approach to detect the particular days during the mobile app maintenance phase when the negative reviews require developers’ attention. Furthermore, the method shall facilitate the mapping of the identified abnormal days towards the updates that result in such negativity in reviews. The method’s purpose is to enable app developers to respond swiftly to significant flaws reflected by user reviews in order to prevent user churns. Full article

The paper deals with learning objects for introductory modeling of simple dynamical systems and design of controllers with disturbance observer-based integral action. They can be used to illustrate the design process of state controllers, state and disturbance observers, as well as to get in touch with a popular approach called active disturbance rejection control (ADRC). In both cases, external disturbances along with errors resulting from model inaccuracies are reconstructed using the extended state observer (ESO). In addition to simulation and design of regulators and observers in Matlab/Simulink, the learning objects focus on the development of relevant terminology and competencies in the field of simulation, modeling and experimentation in traditional laboratories, with the support of LMS Moodle and remote control. The main innovativeness of the paper is to clarify the relationship of ADRC to the traditional state space controllers design and modeling by using two types of linear models approximating controlled nonlinear systems. Full article

This paper presents the initial development of a non-invasive system for identification of the pulse pressure waveform to be used for screening cardiac problems. The system employs a tonometric method using an off-the-shelf force sensor and custom-designed electronic circuits. Initial results on the use of the system are presented. Full article

Alleviating traffic congestion is one of the main challenges for the Internet of Vehicles (IoV) in smart cities. Many congestion pricing systems have been proposed recently. However, most of them focus on punishing the vehicles that use certain roads during peak hours, neglecting the proven fact that rewards can encourage drivers to follow the rules. Therefore, in this paper, we propose a new congestion pricing system based on reward and punishment policies for the IoV in a smart city environment, where the vehicles are rewarded for voluntarily choosing to take an alternative path to alleviate traffic congestion. The proposed system is implemented using vehicular ad hoc networks, which eliminate the need for installing a costly electronic toll collection system. We propose a new virtual currency called T-Coin (traffic coin), that is used to reward the vehicles for their positive attitude. T-Coin is also used in the tender between vehicles to manage the road reservation process. The proposed system uses dynamic pricing to adapt to peak-hour traffic congestion. Using simulated traffic on a real map of Beijing city, we prove the usefulness of T-Coin as a traffic congestion pricing system. Full article

With the rapid development of modern society, generated data has increased exponentially. Finding required data from this huge data pool is an urgent problem that needs to be solved. Hashing technology is widely used in similarity searches of large-scale data. Among them, the ranking-based hashing algorithm has been widely studied due to its accuracy and speed regarding the search results. At present, most ranking-based hashing algorithms construct loss functions by comparing the rank consistency of data in Euclidean and Hamming spaces. However, most of them have high time complexity and long training times, meaning they cannot meet requirements. In order to solve these problems, this paper introduces a distributed Spark framework and implements the ranking-based hashing algorithm in a parallel environment on multiple machines. The experimental results show that the Spark-RLSH (Ranking Listwise Supervision Hashing) can greatly reduce the training time and improve the training efficiency compared with other ranking-based hashing algorithms. Full article

Robotic and other autonomous systems that need to operate in environments with people should respect social rules. One important aspect of this is personal space, which is the space surrounding a person. When two people are in a conversation, they position themselves such that a so-called shared interaction space is created in the middle of them. The aim of the current research was to experimentally investigate the shape and size of this shared interaction space in different formations. In three experiments, we had a robot approaching two people who were having a conversation from 5 different directions, and those people indicated what would be a comfortable distance for the robot to stop. We expected that people would take the personal space of their conversation partner into account when stopping the robot. Findings of the three studies mostly confirm this expectation. Apart from some exceptions, people tend to stop the robot at a similar distance from their conversation partner as from themselves. If these findings are applied in the behavior of robotic and other autonomous systems, people would be more likely to trust and later accept robots in their physical space, ultimately creating natural social interactions between humans and robots. Full article

This paper is a part of an exploratory study with the goal of investigating the applicability of collaboratively-developed enterprise resource planning (the CD-ERP approach) within Libyan public universities. Since it is being proposed in order to overcome the topical issue of information system (IS) development in Libyan higher education (LHE), Libyan universities were investigated to assess IS performance while using the approach of methodological assessment. Accordingly, three public universities were chosen to be studied, namely the University of Tripoli (UOT), Misurata University (MU), and Sirte University (SU). The UOT was run as an initial study to identify practical problems in conducting a formal study. Previous papers have discussed the use of a formal study, in which a methodological assessment of IS performance at Libyan universities was conducted. The findings from all three cases are discussed and analyzed on the basis of selected techniques and models. This analysis indicates the low level of the ISs implemented in LHE at the university level, as well as the lack of system development capabilities within these universities. This paper focuses on the management of information flow between universities during student exchange. Business process maps are presented, and improvements based on the CD-ERP approach are proposed. Full article

Most of the existing research papers study the emotion recognition of Minnan songs from the perspectives of music analysis theory and music appreciation. However, these investigations do not explore any possibility of carrying out an automatic emotion recognition of Minnan songs. In this paper, we propose a model that consists of four main modules to classify the emotion of Minnan songs by using the bimodal data—song lyrics and audio. In the proposed model, an attention-based Long Short-Term Memory (LSTM) neural network is applied to extract lyrical features, and a Convolutional Neural Network (CNN) is used to extract the audio features from the spectrum. Then, two kinds of extracted features are concatenated by multimodal compact bilinear pooling, and finally, the concatenated features are input to the classifying module to determine the song emotion. We designed three experiment groups to investigate the classifying performance of combinations of the four main parts, the comparisons of proposed model with the current approaches and the influence of a few key parameters on the performance of emotion recognition. The results show that the proposed model exhibits better performance over all other experimental groups. The accuracy, precision and recall of the proposed model exceed 0.80 in a combination of appropriate parameters. Full article