This study will identify the superior model by grade recognition of building images of office buildings and residential buildings by two image recognition models that are known to be highly accurate. With respect to expert grade evaluation in facade images of buildings, for practical use in the real estate business, focus on the layer depth of convolutional neural networks (CNN), an AI image recognition technology, and build two models using VGG-16 and ResNet-50, which are different layer depths. Modify the network so that it can be applied to the model, perform transfer learning, and obtain a transition graph of classification accuracy. The results of this accuracy comparison considered showed that VGG-16 outputs higher accuracy than ResNet-50 for both applications, indicating that VGG-16 has the advantage. This also suggests that the performance required of the model (architecture) is not only based on layer thickness, but also includes the need to avoid overlearning due to layer depth.

The organization rate of labor unions, which have supported the development of Japanese industry, has been declining year by year, and this trend continues. In addition, Japanese labor unions are organized by company, and employees are required to serve as union officers, making it difficult to appoint officers. This time, with the cooperation of some labor union organizations, we would like to use network analysis techniques to consider measures to revitalize labor union activities from the perspective of executive relationships.

The demand for energy conservation in large-scale facilities is increasing year by year, and there is a growing momentum for air-conditioning equipment operation managers to respond to this demand. Due to the fear of complaints and the resulting busyness of dealing with them, it becomes difficult for operation managers to make drastic operational changes, to try and implement measures such as relaxing temperature settings or stopping AHU fans. In this study, the purpose of this study is to use the physical quantities related to AHU in central air conditioning equipment to clarify the physical quantities when complaints about the thermal environment arise, and to determine whether the operation manager should relax the set temperature or stop the operation of the AHU fan. Consider the ''discomfort index'' that can be used as a guideline.

Currently, the emergence of technologies such as generative AI, Web3, and other technologies that are significantly changing the existing framework of work requires the need for reskilling in the entire country. However, in many companies and employees, productivity remains low due to the inability to adapt to changes in technology and society. In this study, we will simulate how companies can catch up in a situation where completely new skills are required by society as a whole. We will then conduct a qualitative comparative analysis of the changes in the process caused by differences in the organizational structure of companies and the attributes of their employees, in order to obtain suggestions on how to realize reskilling in organization.

Global warming has caused a number of serious disasters in Japan and abroad. Meanwhile, disaster forecast information is becoming more sophisticated every year, and the means of communication is evolving with the spread of smartphones. Despite the fact that disaster information is readily available, the spread of damage has not been contained. Minimization of damage could be achieved through evacuation actions by those who do not evacuate. The purpose of this study is to derive the best information transmission method for guiding evacuation in river floods. What information communication methods can be considered to achieve effective evacuation? With the goal of changing the behavior of residents, including those who cannot obtain disaster information, those who obtain disaster information but do not take evacuation action, and those who do take evacuation action, we will conduct research on disincentive to evacuation action and intervention methods.

We conducted text mining on data from review sites where information and reputations about companies can be shared, and evaluated the labor environment score from that data through sentiment analysis. As a result of investigating the impact of the labor environment score on corporate performance, a positive correlation was shown between the labor environment score and the operating profit margin. Furthermore, by conducting morphological analysis of the review data and analyzing co-occurring keywords, we were able to identify common words between companies and visually confirm them.

Recently, the importance of ethical, legal, and social issues (ELSI) brought about by science and technology has been intensively discussed, and the social inclusion of handicapped people is the one important issue of the ELSI. Responding to the global environment and aging society, which previously required a cost burden, is now becoming an industry that drives innovation. Similarly, it is foreseeable that the social inclusion of handicapped people, which is recognized as a issue necessary of cost-burden, will drive innovation. In this study, we will use special subsidiaries of social welfare corporations as case studies to investigate and analyze the relationship between the social inclusion of people with disabilities and innovation, and to design social systems that will make social inclusion of people with disabilities sustainable.

Financial news is undeniably crucial for investment decisions, it is considered an effective way to predict stock movement in the natural language processing (NLP) field as well. Since financial news corpus always comes with very few features and lots of noise, the models must be capable of handling ultra-long texts and being strong in feature extraction. The mainstream NLP patterns are generally based on pre-trained language models (PLMs), but PLMs are not good at processing ultra-long texts, so will the PLMs outperform the traditional statistical models in this task? We built several typical NLP patterns for experimental comparison and proposed a text extraction algorithm to improve the ultra-long text handling problem. According to the results, the PLMs have no significant advantage in analyzing ultra-long news corpus, and the algorithm we proposed can improve the accuracy of the PLMs by about 3%.

This study examines the involvement of local communities in addressing traffic accident issues, based on the traffic accident statistics provided by the National Police Agency. Both the frequency of traffic accidents and the resulting injuries have shown a declining trend. Nevertheless, a challenge persists in reducing accidents influenced by contemporary factors, including those involving elderly individuals, young people, and bicycle-related incidents. Given the significant influence and role that local communities are believed to play in resolving this issue, we explore the relationship between them.

The economic impact of mega sporting events including both tangible and intangible legacy is a complex issue and challenging to predict due to the difficulty of accurately estimating. However, traditional methods often encounter problems handling large and complex datasets, noisy input sequences, and missing values. Consequently, this study introduces a novel approach using machine learning to optimize the selection of key indicators, thereby enhancing the predictive accuracy for the multifaceted impacts associated with hosting sporting events, such as the FIFA World Cup. Specifically, we utilize the Boosting Algorithms to analyze the most predictive economic and sportive indicators. As a result, we have developed an economic model for estimating the event's impact, offering a robust analytical tool for policymakers and financial analysts.

The purpose of this study is to investigate the importance of location information and patient volume estimation as strategic factors for clinic management in Japan's increasingly competitive medical institution environment. Official patient surveys are typically used to estimate the number of patients, but this provides only limited information. In this study, we will use urology-related diseases in the elderly as an example, estimate the number of patients based on out-of-hospital prescription dispensing data for each city, ward, town and village, and use synthetic population data to explore the relationship between family structure and diseases. In addition, to evaluate the attractiveness of clinics, we will use GIS to visualize location information and analyze it in combination with location information of competing clinics. Furthermore, we will use the Huff model to search for attractive locations for opening clinics, and for this purpose, we will conduct a questionnaire survey among urological disease patients living in the target area and set numerical values for attractiveness. This is expected to provide strategic geographic information for clinic management.