A graph convolutional generative adversarial network (GCGAN) is proposed to provide recommendations for new users or items. To maintain scalability, the discriminator was improved to capture the latent features of users and items, using graph convolution from a minibatch-sized bipartite graph. In the experiment using MovieLens, it was confirmed that the proposed GCGAN had better performance than the conventional CFGAN, when MovieLens 1M was employed with sufficient data. The proposed method is characterized in such a manner that it can learn domain information of both, users and items, and it does not require to relearn a model for a new node. Further, it can be developed for any service having such conditions, in the information recommendation field.

This paper proposes an optimal design of network and a routing mechanism that is independent from fixed protocols. It provides an optimized route for diversified mesh network, which can support interorganizational communication in a large-scale operation. Decentralization of the system ensures that every protocol acts independently and selects the best optimal path during transmission of data without modifying their architecture and technology. Incorporation of definite source configuration improves the mobility of the systems. Each sensor is individually processed to balance the data load and prevent congestion. Simple transmit-receive test is performed by circulating messages of increasing size between end sensors and network destination. The proposed technique is considered to be effective in terms of interoperability speed, data accuracy and bit error rate (BER) with an increment of 27.13%, 99.98%, and 15.12%, respectively. Finally, the test demonstrates its expediency in terms of adaptability and scalability.

Large annotated datasets are crucial for training deep machine learning models, but they are expensive and time-consuming to create. There are already numerous public datasets, but a vast amount of unlabeled data, especially video data, can still be annotated and leveraged to further improve the performance and accuracy of machine learning models. Therefore, it is essential to reduce the time and effort required to annotate a dataset to prevent bottlenecks in the development of this field. In this study, we propose Anno-Mate, a pair of features integrated into the Computer Vision Annotation Tool (CVAT). It facilitates human–machine collaboration and reduces the required human effort. Anno-Mate comprises Auto-Fit, which uses an EfficientDet-D0 backbone to tighten an existing bounding box around an object, and AutoTrack, which uses a channel and spatial reliability tracking (CSRT) tracker to draw a bounding box on the target object as it moves through the video frames. Both features exhibit a good speed and accuracy trade-off. Auto-Fit garnered an overall accuracy of 87% and an average processing time of 0.47 s, whereas the AutoTrack feature exhibited an overall accuracy of 74.29% and could process 18.54 frames per second. When combined, these features are proven to reduce the time required to annotate a minute of video by 26.56%.

This paper presents a prototype of a centralized contactless traffic violation apprehension system composed of an artificial intelligence (AI) engine and a web application. The AI engine collects traffic data, primarily traffic violation data, through a contactless approach by using different video and image processing techniques and AI algorithms in its three modules: license plate detection, optical character recognition (OCR), and number coding violation detection. The web application consolidates all the data produced by the AI engine and provides a graphical user interface (GUI) for data management, visualization, and analysis. This contactless apprehension system aims to automate, standardize, and streamline the existing processes of law enforcement agencies and institutions for a more efficient apprehension of traffic violators and help them improve their traffic planning and management in the congested areas of the Philippines.

Most automatic license-plate recognition (ALPR) systems use still images and ignore the temporal information in videos. Videos provide rich temporal and motion information that should be considered during training and testing. This study focuses on creating an ALPR system that uses videos. The proposed system is comprised of detection, tracking, and recognition modules. The system achieved accuracies of 81.473% and 84.237% for license-plate detection and classification, respectively.

In recent years, aging of the population has become a major social problem. Various types of sensor networks have been applied to elderly monitoring for addressing the care problems of the elderly living alone. We have also proposed elderly monitoring systems based on wireless sensor network devices. However, vision-based sensors can also cause a mental burden on the elderly’s privacy. Furthermore, the number of sensors must be reduced, if possible. Therefore, this study proposes an elderly monitoring system composed of two vibration sensors placed on the floor and a pneumatic sensor placed on the bed. Because both sensors include considerable measurement noise, we propose a human behavior estimation method that includes anomaly detection from time-series measurement data using an autocorrelation coefficient. Finally, we discuss the effectiveness and usability of the proposed system through several experimental results. The accuracy of walking detection reaches 94.4%, while the error of heartbeat detection is 3.01 bpm.

Human life expectancy is at present the maximum in recorded history. However, a disadvantage is that the elderly are increasingly displaying cognitive disabilities. Studies have shown that physical exercises such as calisthenics can potentially prevent disabilities. Meanwhile, existing systems for evaluating human pose focus mainly on accuracy and omit convenience and efficiency. To solve this issue, in this paper, we propose a framework for rapidly estimating three-dimensional human pose from two camera views. It is based on an evolutionary algorithm. This system can be applied straightforwardly to inexpensive smart devices and used to evaluate multiple individuals’ calisthenics with two or more smart devices.

In this paper, an improved long short-term memory (LSTM)-based deep neural network structure is proposed for learning variable-length Chinese sentence semantic similarities. Siamese LSTM, a sequence-insensitive deep neural network model, has a limited ability to capture the semantics of natural language because it has difficulty explaining semantic differences based on the differences in syntactic structures or word order in a sentence. Therefore, the proposed model integrates the syntactic component features of the words in the sentence into a word vector representation layer to express the syntactic structure information of the sentence and the interdependence between words. Moreover, a relative position embedding layer is introduced into the model, and the relative position of the words in the sentence is mapped to a high-dimensional space to capture the local position information of the words. With this model, a parallel structure is used to map two sentences into the same high-dimensional space to obtain a fixed-length sentence vector representation. After aggregation, the sentence similarity is computed in the output layer. Experiments with Chinese sentences show that the model can achieve good results in the calculation of the semantic similarity.

The cortical learning algorithm (CLA) is a time-series data prediction method that is designed based on the human neocortex. The CLA has multiple columns that are associated with the input data bits by synapses. The input data is then converted into an internal column representation based on the synapse relation. Because the synapse relation between the columns and input data bits is fixed during the entire prediction process in the conventional CLA, it cannot adapt to input data biases. Consequently, columns not used for internal representations arise, resulting in a low prediction accuracy in the conventional CLA. To improve the prediction accuracy of the CLA, we propose a CLA that self-adaptively arranges the column synapses according to the input data tendencies and verify its effectiveness with several artificial time-series data and real-world electricity load prediction data from New York City. Experimental results show that the proposed CLA achieves higher prediction accuracy than the conventional CLA and LSTMs with different network optimization algorithms by arranging column synapses according to the input data tendency.

With the recent developments in robotics, the ability of robots to recognize their environment has significantly improved. However, the manner in which robots behave depending on a particular situation remains an unsolved problem. In this study, we propose a decision-making method for robots based on robot ethics. Specifically, we applied the two-level theory of utilitarianism, comprising SYSTEM 1 (intuitive level) for quick decisions and SYSTEM 2 (critical level) for slow but careful decisions. SYSTEM 1 represented a set of heuristically determined responses and SYSTEM 2 represented a rule-based discriminator. The decision-making method was as follows. First, SYSTEM 1 selected the response to the input. Next, SYSTEM 2 selected the rule that the robot’s behavior should follow depending on the amount of happiness and unhappiness of the human, robot, situation, and society. We assumed three choices for SYSTEM 2. We assigned “non-cooperation” to asocial comments, “cooperation” to when the amount of happiness was considered to be high beyond the status quo bias, and “withholding” to all other cases. In the case of choosing between cooperation or non-cooperation, we modified the behavior selected in SYSTEM 1. An impression evaluation experiment was conducted, and the effectiveness of the proposed method was demonstrated.

Three-mode fuzzy co-clustering is a promising technique for analyzing relational co-occurrence information among three mode elements. The conventional FCM-type algorithms achieved simultaneous fuzzy partition of three mode elements based on the fuzzy c-means (FCM) concept, and then, they often suffer from careful tuning of three independent fuzzification parameters. In this paper, a novel three-mode fuzzy co-clustering algorithm is proposed by modifying the conventional aggregation criterion of three elements based on a probabilistic concept. The fuzziness degree of three-mode partition can be easily tuned only with a single parameter under the guideline of the probabilistic standard. The characteristic features of the proposed method are compared with the conventional algorithms through numerical experiments using an artificial dataset and are demonstrated in application to a real world dataset of MovieLens movie evaluation data.

Aging has become a global social issue nowadays. We want to provide an elderly care system for older people who live alone. Based on the perspective of an informationally structured space (ISS), we have developed a monitoring system by using high-precision vibration sensors. In preliminary experiments, we observed that the autocorrelation coefficient reflected periodic human activities to a certain extent. Therefore, we propose a time delay neural network (TDNN) with autocorrelation as the input to analyze the vibration data. The system can estimate the current state of the elderly. When the system observes any abnormal situation of the elderly, the system can confirm by voice or notify the caregiver, if necessary. In the experiments, we compared the proposed method with traditional TDNNs using raw data as the input. The results demonstrated that proposed methods had performed well when using vibration sensors to measure user behaviors in the bathroom and living room.

This paper proposes Treemap-based visualization for supporting cluster analysis of multi-dimensional data. It is important to grasp data distribution in a target dataset for such tasks as machine learning and cluster analysis. When dealing with multi-dimensional data such as statistical data and document datasets, dimensionality reduction algorithms are usually applied to project original data to lower-dimensional space. However, dimensionality reduction tends to lose the characteristics of data in the original space. In particular, the border between different data groups could not be represented correctly in lower-dimensional space. To overcome this problem, the proposed visualization method applies Fuzzy c-Means to target data and visualizes the result on the basis of the highest and the second-highest membership values with Treemap. Visualizing the information about not only the closest clusters but also the second closest ones is expected to be useful for identifying objects around the border between different clusters, as well as for understanding the relationship between different clusters. A prototype interface is implemented, of which the effectiveness is investigated with a user experiment on a news articles dataset. As another kind of text data, a case study of applying it to a word embedding space is also shown.