Kiyoshi Takami

Sung-Yoo Lim, Koki Sato, Kiyoshi Takami et al.

Social activities result from complex joint activity-travel decisions between group members. While observing the decision-making process of these activities is difficult via traditional travel surveys, the advent of new types of data, such as unstructured chat data, can help shed some light on these complex processes. However, interpreting these decision-making processes requires inferring both explicit and implicit factors. This typically involves the labor-intensive task of manually annotating dialogues to capture context-dependent meanings shaped by the social and cultural norms. This study evaluates the potential of Large Language Models (LLMs) to automate and complement human annotation in interpreting decision-making processes from group chats, using data on joint eating-out activities in Japan as a case study. We designed a prompting framework inspired by the knowledge acquisition process, which sequentially extracts key decision-making factors, including the group-level restaurant choice set and outcome, individual preferences of each alternative, and the specific attributes driving those preferences. This structured process guides the LLM to interpret group chat data, converting unstructured dialogues into structured tabular data describing decision-making factors. To evaluate LLM-driven outputs, we conduct a quantitative analysis using a human-annotated ground truth dataset and a qualitative error analysis to examine model limitations. Results show that while the LLM reliably captures explicit decision-making factors, it struggles to identify nuanced implicit factors that human annotators readily identified. We pinpoint specific contexts when LLM-based extraction can be trusted versus when human oversight remains essential. These findings highlight both the potential and limitations of LLM-based analysis for incorporating non-traditional data sources on social activities.

Itsuki Sato, Kiyoshi Takami, Giancarlos Parady

Realistic assessments of school commuting accessibility in areas with infrequent public transport services require accounting for operational delays; however, the impact of these delays has not been sufficiently examined. This study evaluates high-school accessibility in Matsumoto City, a regional city in Japan, using GTFS data representing both scheduled timetables and actual operating conditions. Accessibility levels are assessed under scheduled operations, while the effects of delays are examined through a comparative analysis based on actual delay measurements over a five-day workweek. Furthermore, a sensitivity analysis of travel-time thresholds was conducted. Results show that, when walking, cycling to stations, and public transport use are allowed, 78% of children under 15 can reach at least one high school within a 90-minute round trip, and 67% within a 60-minute round trip. Extending the threshold to 120 minutes enables access to nearly all schools in the city center, but the overall proportion increases only marginally to 81%. Delay impacts are particularly pronounced along bus routes connecting the central station with suburban areas, while in some areas, delays generate idiosyncratic events, where irregular transfers and reduced waiting times result in improved accessibility. Results underscore the need for both short-term measures,such as adjusting school start times, prioritizing buses, and introducing dedicated school routes, and long-term strategies, such as incorporating public transport accessibility into school consolidation decisions, to guarantee fair access to education opportunities without relying on private vehicles.