A Bottom-Up Modeling Methodology Using Knowledge Graphs for Composite Metric Development Applied to Traffic Crashes in the State of Texas
Data is a key factor for understanding real-world phenomena. Data can be discovered and integrated from multiple sources and has the potential to be interpreted in a multitude of ways. Traffic crashes, for example, are common events that occur in cities and provide a significant amount of data that has potential to be analyzed and disseminated in a way that can improve mobility of people, and ultimately improve the quality of life. Improving the quality of life of city residents through the use of data and technology is at the core of Smart Cities solutions. Measuring the improvement that Smart Cities provide usually relies on data collection and analytics before and after the implementation of such solutions. Through a methodological approach, implicit information about mobility, in particular traffic crash data, can be discovered and used in the interpretation and dissemination of information through different data views, such as metrics and narratives, thus fostering the gain of knowledge. In this work, a novel modeling methodology for traffic crashes was developed, namely the Bottom-Up Modeling (BUM) methodology. This methodology integrates publicly available mobility data, proposes a data model implemented in a knowledge graph that includes semantic annotations, and produces a composite metric called the Critical Composite Index (CCI). The CCI uses weighted criteria values to make each crash comparable to others with similar data provided. The BUM methodology was applied to model traffic crashes between different geographic locations in Texas. The resulting methodology enables the creation of metrics for use by many stakeholders, particularly non-domain experts. The use of the BUM methodology to generate different perspectives of the crash data is addressed through the generation of different data views (i.e., metrics and data narratives). Moreover, an ontology was developed based on the knowledge graph to formalize the proposed data model to, verify logic consistency and infer implicit information with the use of a generic description logic reasoner. The CCI metric was evaluated by comparing against currently used frequency metrics and a user-evaluation survey. Evaluation results show that the CCI provides improved knowledge gain over currently used metrics. This work contributes to data science research, using an interdisciplinary approach that involves Computer Science techniques, mathematics and domain expertise to address complex challenges, such as those in converting cities to Smart Cities.
Computer science|Artificial intelligence|Transportation
Mejia, Daniel Michael, "A Bottom-Up Modeling Methodology Using Knowledge Graphs for Composite Metric Development Applied to Traffic Crashes in the State of Texas" (2019). ETD Collection for University of Texas, El Paso. AAI13881043.