{"id":{"repo_id":"vt","oai_identifier":"oai:vtechworks.lib.vt.edu:10919/135057"},"canonical_url":"https://search.dev.ndltd.org/etd/vt/oai:vtechworks.lib.vt.edu:10919/135057","repository":{"repo_id":"vt","name":"Virginia Tech","base_url":"https://vtechworks.lib.vt.edu/oai/request"},"display":{"title":"Assessing Bikeability in Virginia: A Comparison of CHATGPT-4o and Traditional Models","abstract":"This study evaluates cycling friendliness by using Street View Images (SVIs) and deep learning models to assess street-level features associated with the cycling environment in Virginia. An image segmentation model, PixelLib, was used to extract the proportion of seven features, namely greenery, streetlights, roads, sidewalks, cars, sky, and buildings, from 1,727 Google Street View Images (SVIs). The estimated features were also used to calculate the scene complexity level (CL), reflecting the visual navigability of each location. Results revealed that many SVIs exhibit a moderate CL, indicating that the cycling environment offers clear visual conditions that may improve safety for cyclists. The CL, combined with three other factors: greenery, traffic volume (Average Annual Daily Traffic), and bike lane types, were used to develop the Bikeability Index (BI). The distribution of the BI varied across Virginia, with classes ranging from low to very high. Higher BI pointed to more bicycle-friendly environments, while lower BI indicated areas with conditions less supportive of cycling. The study also demonstrated the capability of a large language model (LLM), GPT-4o, to support transportation analysis through the automated classification of bike lane infrastructure. The model achieved a classification accuracy of 98.28% when used to categorize 294 SVIs into separated and shared bike lanes. In addition, GPT-4o was used to generate bikeabillity rating for ten SVIs based on a structured prompt. A correlation analysis revealed a weak but statistically significant negative relationship between the BI and the CDC's Social Vulnerability Index (SoVI), suggesting that socially vulnerable areas tend to have less bikeable environments.","abstract_html":"This study evaluates cycling friendliness by using Street View Images (SVIs) and deep learning models to assess street-level features associated with the cycling environment in Virginia. An image segmentation model, PixelLib, was used to extract the proportion of seven features, namely greenery, streetlights, roads, sidewalks, cars, sky, and buildings, from 1,727 Google Street View Images (SVIs). The estimated features were also used to calculate the scene complexity level (CL), reflecting the visual navigability of each location. Results revealed that many SVIs exhibit a moderate CL, indicating that the cycling environment offers clear visual conditions that may improve safety for cyclists. The CL, combined with three other factors: greenery, traffic volume (Average Annual Daily Traffic), and bike lane types, were used to develop the Bikeability Index (BI). The distribution of the BI varied across Virginia, with classes ranging from low to very high. Higher BI pointed to more bicycle-friendly environments, while lower BI indicated areas with conditions less supportive of cycling. The study also demonstrated the capability of a large language model (LLM), GPT-4o, to support transportation analysis through the automated classification of bike lane infrastructure. The model achieved a classification accuracy of 98.28% when used to categorize 294 SVIs into separated and shared bike lanes. In addition, GPT-4o was used to generate bikeabillity rating for ten SVIs based on a structured prompt. A correlation analysis revealed a weak but statistically significant negative relationship between the BI and the CDC's Social Vulnerability Index (SoVI), suggesting that socially vulnerable areas tend to have less bikeable environments.","abstract_has_math":false,"creators":["Lawal, Abdul-Azeez Ademola"],"institution":"Virginia Tech","degree_name":"Master of Science","degree_level":"masters","degree_discipline":"Geography","degree_department":"Geography","school":null,"contributors":[],"advisors":[],"committee_chairs":["Kim, Junghwan"],"committee_members":["Crawford, Thomas Wall","Rijal, Santosh"],"year":2025,"date_issued":"2025-06-04","date_published":"2025-06-04","updated_at":"2026-07-22T22:19:48Z","subjects":["Bikeability","CHATGPT","Deep Learning","Street View Images","Non-Motorized Transportation","Geography","Computer Vision"],"languages":["en"],"rights":["In Copyright"],"rights_urls":["http://rightsstatements.org/vocab/InC/1.0/"],"identifier_entries":[{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["vt_gsexam:43231"],"render_values":[{"text":"vt_gsexam:43231","href":null,"code":true}]}]},"links":{"outbound_url":"https://hdl.handle.net/10919/135057","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.committeechair","label":"Committee Chair","values":["Kim, Junghwan"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Crawford, Thomas Wall","Rijal, Santosh"]},{"key":"dc:contributor.department","label":"Department","values":["Geography"]},{"key":"dc:creator","label":"Author","values":["Lawal, Abdul-Azeez Ademola"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2025-06-05T08:00:27Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2025-06-05T08:00:27Z"]},{"key":"dc:date.issued","label":"Date","values":["2025-06-04"]},{"key":"dc:publisher","label":"Institution","values":["Virginia Tech"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Geography"]},{"key":"thesis:degree_level","label":"Degree Level","values":["masters"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Virginia Polytechnic Institute and State University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Bikeability","CHATGPT","Deep Learning","Street View Images","Non-Motorized Transportation","Geography","Computer Vision"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["In Copyright"]},{"key":"dc:rights.uri","label":"Rights URI","values":["http://rightsstatements.org/vocab/InC/1.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["vt_gsexam:43231"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/10919/135057"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This study evaluates cycling friendliness by using Street View Images (SVIs) and deep learning models to assess street-level features associated with the cycling environment in Virginia. An image segmentation model, PixelLib, was used to extract the proportion of seven features, namely greenery, streetlights, roads, sidewalks, cars, sky, and buildings, from 1,727 Google Street View Images (SVIs). The estimated features were also used to calculate the scene complexity level (CL), reflecting the visual navigability of each location. Results revealed that many SVIs exhibit a moderate CL, indicating that the cycling environment offers clear visual conditions that may improve safety for cyclists. The CL, combined with three other factors: greenery, traffic volume (Average Annual Daily Traffic), and bike lane types, were used to develop the Bikeability Index (BI). The distribution of the BI varied across Virginia, with classes ranging from low to very high. Higher BI pointed to more bicycle-friendly environments, while lower BI indicated areas with conditions less supportive of cycling. The study also demonstrated the capability of a large language model (LLM), GPT-4o, to support transportation analysis through the automated classification of bike lane infrastructure. The model achieved a classification accuracy of 98.28% when used to categorize 294 SVIs into separated and shared bike lanes. In addition, GPT-4o was used to generate bikeabillity rating for ten SVIs based on a structured prompt. A correlation analysis revealed a weak but statistically significant negative relationship between the BI and the CDC's Social Vulnerability Index (SoVI), suggesting that socially vulnerable areas tend to have less bikeable environments."]},{"key":"dc:description.abstractgeneral","label":"General Abstract","values":["Bicycling has gained increasing popularity as a sustainable mode of transportation, drawing attention to the safety and equity of cycling environments in urban areas. This study utilizes Street View Images (SVIs) and deep learning methods to evaluate the quality of the bicycling environment in Virginia. A multimodal AI model, GPT-4o, was employed with custom-designed prompts for assessing cycling friendliness and bike lane classification of 400 SVIs. The model ranked scenes using objective factors such as greenery, sidewalks, and streetlights, closely aligning with experts' evaluations and achieved a classification accuracy of 98.28%. To complement this assessment, an image segmentation model, PixelLib, was used to extract street-level features. These features were combined with traffic volume and visual complexity to develop a bikeability index (BI), which was then analyzed in relation to the Social Vulnerability Index (SoVI). Results reveal a negative correlation: areas with higher social vulnerability often have lower BI. This research highlights spatial disparities in cycling friendliness and offers a data-driven approach to promoting more equitable and sustainable urban mobility."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["Master of Science"]},{"key":"dc:format.medium","label":"Dc Format Medium","values":["ETD"]},{"key":"dc:title","label":"Title","values":["Assessing Bikeability in Virginia: A Comparison of CHATGPT-4o and Traditional Models"]}]}],"canonical_facts":{"dc:contributor.committeechair":["Kim, Junghwan"],"dc:contributor.committeemember":["Crawford, Thomas Wall","Rijal, Santosh"],"dc:contributor.department":["Geography"],"dc:creator":["Lawal, Abdul-Azeez Ademola"],"dc:date.accessioned":["2025-06-05T08:00:27Z"],"dc:date.available":["2025-06-05T08:00:27Z"],"dc:date.issued":["2025-06-04"],"dc:description.abstract":["This study evaluates cycling friendliness by using Street View Images (SVIs) and deep learning models to assess street-level features associated with the cycling environment in Virginia. An image segmentation model, PixelLib, was used to extract the proportion of seven features, namely greenery, streetlights, roads, sidewalks, cars, sky, and buildings, from 1,727 Google Street View Images (SVIs). The estimated features were also used to calculate the scene complexity level (CL), reflecting the visual navigability of each location. Results revealed that many SVIs exhibit a moderate CL, indicating that the cycling environment offers clear visual conditions that may improve safety for cyclists. The CL, combined with three other factors: greenery, traffic volume (Average Annual Daily Traffic), and bike lane types, were used to develop the Bikeability Index (BI). The distribution of the BI varied across Virginia, with classes ranging from low to very high. Higher BI pointed to more bicycle-friendly environments, while lower BI indicated areas with conditions less supportive of cycling. The study also demonstrated the capability of a large language model (LLM), GPT-4o, to support transportation analysis through the automated classification of bike lane infrastructure. The model achieved a classification accuracy of 98.28% when used to categorize 294 SVIs into separated and shared bike lanes. In addition, GPT-4o was used to generate bikeabillity rating for ten SVIs based on a structured prompt. A correlation analysis revealed a weak but statistically significant negative relationship between the BI and the CDC's Social Vulnerability Index (SoVI), suggesting that socially vulnerable areas tend to have less bikeable environments."],"dc:description.abstractgeneral":["Bicycling has gained increasing popularity as a sustainable mode of transportation, drawing attention to the safety and equity of cycling environments in urban areas. This study utilizes Street View Images (SVIs) and deep learning methods to evaluate the quality of the bicycling environment in Virginia. A multimodal AI model, GPT-4o, was employed with custom-designed prompts for assessing cycling friendliness and bike lane classification of 400 SVIs. The model ranked scenes using objective factors such as greenery, sidewalks, and streetlights, closely aligning with experts' evaluations and achieved a classification accuracy of 98.28%. To complement this assessment, an image segmentation model, PixelLib, was used to extract street-level features. These features were combined with traffic volume and visual complexity to develop a bikeability index (BI), which was then analyzed in relation to the Social Vulnerability Index (SoVI). Results reveal a negative correlation: areas with higher social vulnerability often have lower BI. This research highlights spatial disparities in cycling friendliness and offers a data-driven approach to promoting more equitable and sustainable urban mobility."],"dc:description.degree":["Master of Science"],"dc:format.medium":["ETD"],"dc:identifier.other":["vt_gsexam:43231"],"dc:identifier.uri":["https://hdl.handle.net/10919/135057"],"dc:language.iso":["en"],"dc:publisher":["Virginia Tech"],"dc:rights":["In Copyright"],"dc:rights.uri":["http://rightsstatements.org/vocab/InC/1.0/"],"dc:subject":["Bikeability","CHATGPT","Deep Learning","Street View Images","Non-Motorized Transportation","Geography","Computer Vision"],"dc:title":["Assessing Bikeability in Virginia: A Comparison of CHATGPT-4o and Traditional Models"],"dc:type":["Thesis"],"thesis:degree_discipline":["Geography"],"thesis:degree_level":["masters"],"thesis:degree_name":["Master of Science"],"thesis:institution_name":["Virginia Polytechnic Institute and State University"]},"updated_at":"2026-07-22T22:19:48Z"}