**Features** ■ Unbiasedness: Objective and comprehensive analysis of a vast number of papers AI objectively and comprehensively analyzes a vast amount of paper information without being influenced by researchers' knowledge, interests, or the journals in which they are published. It converts words and sentences into vectors (numerical values) based on relevance and similarity, presenting them as positional information, allowing for a quick grasp of the results at a glance. ■ Serendipity: Increasing the probability of unexpected discoveries Thanks to vector transformation technology, unexpected information and literature that could not be found through keyword searches can also be detected. Searches can exclude unnecessary elements, such as "excluding cancer," making access to other fields and unknown information efficient. ■ Discovery: Gaining new insights By gaining new insights from discoveries and realizations made with KIBIT Amanogawa, you can advance your research and lead to further exploration. ● Trained on all English papers published in PubMed, enabling immediate exploration and analysis ● A cloud service that can be started in as little as about one week from application ● Lightweight processing, eliminating the need for large-scale facilities such as supercomputers or large server clusters

【Function】 Plots papers based on location information and displays them on the "Space Map." Users can select a portion of the map to narrow down their search to only the clusters in that area, and they can also reference abstracts on the system screen. The database of papers is updated every two weeks, and the new paper filtering feature supports users in reliably catching up with the publication status and content of new literature. Search results can be analyzed from various angles, including "Terms Frequency," which shows what each cluster contains, and "Chronological Chart," which illustrates the annual trends of each cluster, allowing for a quick grasp of the trends in paper content. The "Word Cloud" excludes common words such as articles and is composed only of words that have research significance, such as molecular names, disease names, and medical terms, visualizing necessary information without the noise of common words. In the "Card Box," the plotted paper information is displayed in order of relevance and similarity to search terms or hypotheses, and it is also possible to output a list of the top 100 papers in Excel format. The display size and position of each panel can be freely customized by the user according to their preferences and objectives.