{"643350":{"#nid":"643350","#data":{"type":"news","title":"School of Public Policy Study Shows Artificial Intelligence Can Beat Human Experts in Discovering Attitudes on Electric Vehicle Charging","body":[{"value":"\u003Cp\u003EA research team led by \u003Ca href=\u0022https:\/\/spp.gatech.edu\/\u0022\u003ESchool of Public Policy\u003C\/a\u003E Assistant Professor Omar Asensio used artificial intelligence to analyze consumer data from electric vehicle (EV) charging station reviews, besting people who analyzed and annotated the same data. The algorithms also detected social inequities in charging station availability across the United States, according to their \u003Ca href=\u0022https:\/\/doi.org\/10.1016\/j.patter.2020.100195\u0022\u003Eresearch\u003C\/a\u003E published Jan. 22 in the journal \u003Cem\u003EPatterns.\u003C\/em\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026ldquo;By deploying data strategies that can generate real-time insights at different scales of analysis, we\u0026rsquo;ve demonstrated that artificial intelligence can provide inexpensive, highly accurate insights for industry and policymakers seeking to build out the electric vehicle charging infrastructure,\u0026rdquo; Asensio said. \u0026ldquo;This is an important goal given the potential climate-change benefits of emission-free vehicles and related policies.\u0026rdquo;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Ca href=\u0022https:\/\/iac.gatech.edu\/people\/person\/omar-isaac-asensio\u0022\u003EAsensio\u003C\/a\u003E co-authored the \u003Cem\u003EPatterns \u003C\/em\u003Epaper with student researchers Daniel Marchetto, a Ph.D. student in Public Policy; Susie Ha, a dual Ph.D. student in Computational Science and Engineering and Civil and Environmental Engineering; and Sameer Dharur, a master\u0026rsquo;s student in Computer Science.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThe work builds on \u003Ca href=\u0022https:\/\/iac.gatech.edu\/research\/features\/deep-learning-electric-vehicle-charging-research\u0022\u003Eearlier efforts\u003C\/a\u003E by Asensio and his Georgia Tech students using deep learning models to assess consumer sentiment from unstructured text reviews of charging stations submitted to popular smartphone apps for EV drivers.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThat research showed promise using a convolutional neural network to detect consumer sentiment in near real-time. The technique rivaled human performance in classifying sentiment. However, being able to detect the reasons for consumer positivity or dismay remained an open problem.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EIn this new research, however, Asensio\u0026rsquo;s team shows that transformer-based deep learning, a different machine learning technique, was able to discover topics of consumer discussion automatically.\u0026nbsp;In some cases, the models\u0026nbsp;did\u0026nbsp;better than expert human annotators\u0026nbsp;in labeling the conversations.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EFor instance, consider the review, \u0026ldquo;Thanks very busy dealership but happy to allow use of qcdc.\u0026rdquo; The model was able to identify labels accurately representing the reasons for the review: \u003Cem\u003Efunctionality\u003C\/em\u003E, \u003Cem\u003Eavailability\u003C\/em\u003E, and \u0026ldquo;\u003Cem\u003Edealership\u003C\/em\u003E.\u0026rdquo; Asensio\u0026rsquo;s team used the latter label to describe comments concerning specific dealerships and users\u0026rsquo; associated charging experiences.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EPeople did not do quite as well as the transformer-based BERT (bidirectional encoder representations from transformers) model in applying the correct labels, according to the research. The BERT model applied the correct labels to reviews 91.6% of the time. That model and another used by researchers, XLNet, in some cases performed 3- to 5-percentage points better than human experts.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026ldquo;We are at a new\u0026nbsp;milestone\u0026nbsp;of research when AI beats humans, especially in domains where cheap crowd-sourced labels are difficult and costly to get,\u0026rdquo; Asensio said.\u003C\/p\u003E\r\n\r\n\u003Ch2\u003EAnalysis Reveals Inequities\u003C\/h2\u003E\r\n\r\n\u003Cp\u003EIn addition to the classification advances, Asensio\u0026rsquo;s team spotted a pattern of station availability issues from existing EV users. The study merged the AI model predictions with location features to evaluate possible differences by region. The study found evidence that consumers report station availability issues most frequently in smaller cities, particularly in the West, Midwest, and Hawaii.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThe fact that station availability issues tend to dominate in smaller communities with 10,000 to 50,000 residents suggests that additional mechanisms are needed to broaden access to sustainable charging infrastructure, Asensio said.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThese findings are notable because the United States lacks federal innovation policies that could address these potential gaps, which Asensio said result from a decentralized model of growth.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026ldquo;The latest theory and evidence suggest that it may be more effective to subsidize EV station deployment, rather than subsidize EV car sales. However, until the advent of these AI-driven tools, we didn\u0026rsquo;t have a practical way to discover large-scale behavioral issues,\u0026rdquo; Asensio said.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThe team is now rapidly accelerating efforts to better understand equity issues in the availability of station infrastructure for hard-to-reach communities, especially considering the current rapid global investment in the area.\u003C\/p\u003E\r\n\r\n\u003Ch2\u003EAdvances Could Fuel Policy Innovation\u003C\/h2\u003E\r\n\r\n\u003Cp\u003EThe advances demonstrated in Asensio\u0026rsquo;s paper buttress the case that machine learning tools could uniquely address the need for real-time consumer intelligence related to electric mobility.\u003C\/p\u003E\r\n\r\n\u003Cp\u003ESuch improvements are significant because of the unique role charging infrastructure plays in supporting the sale of electric vehicles, an expanding market seen as a critical tool to help reduce greenhouse gas emissions and combat climate change.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThe notable advances also have broader implications for public policy work in general, according to Asensio and his team.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026ldquo;The extent of this improvement could significantly accelerate automated research evaluation using large-scale consumer data for performance assessment and regional policy analysis in other domains,\u0026rdquo; the team wrote in the paper.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThe article, \u0026ldquo;Topic Classification of Electric Vehicle Consumer Experiences with Transformer-Based Deep Learning,\u0026rdquo; is available online at \u003Ca href=\u0022https:\/\/doi.org\/10.1016\/j.patter.2020.100195\u0022\u003Ehttps:\/\/doi.org\/10.1016\/j.patter.2020.100195\u003C\/a\u003E.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThe School of Public Policy is a unit of the \u003Ca href=\u0022https:\/\/iac.gatech.edu\u0022\u003EIvan Allen College of Liberal Arts\u003C\/a\u003E.\u003C\/p\u003E\r\n","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp\u003EThe research demonstrates advances in using artifical intelligence for policy analysis, and reveals\u0026nbsp;social inequities in charging station availability across the United States,\u003C\/p\u003E\r\n","format":"limited_html"}],"field_summary_sentence":[{"value":"The research demonstrates advances in using artificial intelligence for policy analysis."}],"uid":"34600","created_gmt":"2021-01-22 17:45:25","changed_gmt":"2021-01-22 17:45:52","author":"mpearson34","boilerplate_text":"","field_publication":"","field_article_url":"","dateline":{"date":"2021-01-22T00:00:00-05:00","iso_date":"2021-01-22T00:00:00-05:00","tz":"America\/New_York"},"extras":[],"hg_media":{"643349":{"id":"643349","type":"image","title":"EV charging study students","body":null,"created":"1611337385","gmt_created":"2021-01-22 17:43:05","changed":"1611337385","gmt_changed":"2021-01-22 17:43:05","alt":"Students Daniel Marchetto, Sooji (Susie) Ha, and\u00a0Sameer Dharur","file":{"fid":"244265","name":"ev charging students.jpg","image_path":"\/sites\/default\/files\/images\/ev%20charging%20students_0.jpg","image_full_path":"http:\/\/tlwarc.hg.gatech.edu\/\/sites\/default\/files\/images\/ev%20charging%20students_0.jpg","mime":"image\/jpeg","size":308032,"path_740":"http:\/\/tlwarc.hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/images\/ev%20charging%20students_0.jpg?itok=32_wkmBS"}},"639033":{"id":"639033","type":"image","title":"Omar Isaac Asensio, Assistant Professor ","body":null,"created":"1600090403","gmt_created":"2020-09-14 13:33:23","changed":"1600090403","gmt_changed":"2020-09-14 13:33:23","alt":"Omar Isaac Asensio, Assistant Professor\u00a0","file":{"fid":"242970","name":"Omar Asensio-Assistant Professor-v2.jpg","image_path":"\/sites\/default\/files\/images\/Omar%20Asensio-Assistant%20Professor-v2.jpg","image_full_path":"http:\/\/tlwarc.hg.gatech.edu\/\/sites\/default\/files\/images\/Omar%20Asensio-Assistant%20Professor-v2.jpg","mime":"image\/jpeg","size":104509,"path_740":"http:\/\/tlwarc.hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/images\/Omar%20Asensio-Assistant%20Professor-v2.jpg?itok=4aPO6MNQ"}}},"media_ids":["643349","639033"],"groups":[{"id":"1281","name":"Ivan Allen College of Liberal Arts"},{"id":"1289","name":"School of Public Policy"}],"categories":[{"id":"142","name":"City Planning, Transportation, and Urban Growth"},{"id":"151","name":"Policy, Social Sciences, and Liberal Arts"}],"keywords":[],"core_research_areas":[{"id":"39511","name":"Public Service, Leadership, and Policy"}],"news_room_topics":[],"event_categories":[],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp\u003EMichael Pearson\u003Cbr \/\u003E\r\nmichael.pearson@iac.gatech.edu\u003C\/p\u003E\r\n","format":"limited_html"}],"email":["michael.pearson@iac.gatech.edu"],"slides":[],"orientation":[],"userdata":""}}}