Populations and land development have the potential to shift as economies change at a rate that is faster than currently employed for updating a transportation plan for a region. This thesis uses the Foursquare location-based social networking check-in data to analyze the origin-destination travel demand for Austin, Texas. A doubly-constrained gravity model has been employed to create an origin-destination model. This model was analyzed in comparison to a singly-constrained gravity model as well as the Capital Area Metropolitan Planning Organization's 2010 Urban Transportation Study's origin-destination matrices through trip length distributions, the zonal origin-destination flow patterns, and the zonal trip generation and attraction heat maps in an effort to validate the methodology. / text

Today’s metropolitan areas see changes in populations and land development occurring at faster rates than transportation planning can be updated. This dissertation explores the use of a new dataset from the location-based social networking spectrum to analyze origin-destination travel demand within Austin, TX. A detailed exploration of the proposed data source is conducted to determine its overall capabilities with respect to the Austin area demographics. A new methodology is proposed for the creation of origin-destination matrices using a peer-to-peer modeling structure. This methodology is compared against a previously examined and more traditional approach, the doubly-constrained gravity model, to understand the capabilities of both models with various friction functions. Each method is examined within the constructs of the study area’s existing origin-destination matrix by examining the coincidence ratios, mean errors, mean absolute errors, frequency ratios, swap ratios, trip length distributions, zonal trip generation and attraction heat maps, and zonal origin-destination flow patterns. Through multiple measures, this dissertation provides initial interpretations of the robust Foursquare data collected for the Austin area. Based upon the data analytics performed, the Foursquare data source is shown to be capable of providing immensely detailed spatial-temporal data that can be utilized as a supplementary data source to traditional transportation planning data collection methods or in conjunction with other data sources, such as social networking platforms. The examination of the proposed peer-to-peer methodology presented within this dissertation provides a first look at the potential of many-to-many modeling for transportation planning. The peer-to-peer model was found to be superior to the doubly-constrained gravity model with respect to intrazonal trips. Furthermore, the peer-to-peer model was found to better estimate productions, attractions, and zone to zone movements when a linear function was used for long trips, and was computationally more proficient for all models examined.

Submitted by Rosivalda Pereira (mrs.pereira@ufma.br) on 2017-08-16T17:52:50Z No. of bitstreams: 1 RodrigoFeitosa.pdf: 4009932 bytes, checksum: 55ef5e97ddc653cf1849e17eafdc396f (MD5) / Made available in DSpace on 2017-08-16T17:52:50Z (GMT). No. of bitstreams: 1 RodrigoFeitosa.pdf: 4009932 bytes, checksum: 55ef5e97ddc653cf1849e17eafdc396f (MD5) Previous issue date: 2013-03-22 / The search of knowledge and its manipulation in companies, institutions or other organizations has become a challenge nowadays. Mostly due to two aspects: the large volume of information available and the difficulty in extracting the knowledge proper to each person (intellectual capital). This difficulty becomes more accentuated when the scenario involved the extraction of knowledge is the Web. The area of Knowledge Management seeks a solution to the limitations described above. Techniques for extracting and control of knowledge can be adopted with the use of Artificial Intelligence, particularly the Knowledge Discovery in Databases. This work proposes the creation of a methodology and application that perform the Data Mining with textual information linked to geo data in a social network, in order to promote Social Recommendation. However, approaches in building recommendation systems present some shortcomings in filtering the results and the way they are suggested to users. The research aims to remedy these deficiencies and addresses issues that still need to search more effective and consolidated results. / A busca do conhecimento e a sua manipulação em empresas, instituições ou outras organizações tem se tornado um desafio nos dias atuais. Em grande parte devido a dois aspectos: o grande volume de informação disponibilizada e a dificuldade em extrair o conhecimento próprio de cada pessoa (capital intelectual). Essa dificuldade torna-se mais acentuada quando o cenário envolvido para a extração de conhecimento é a Web. A área da Gestão de Conhecimento busca a solução para as limitações descritas anteriormente. Técnicas para a extração e controle do conhecimento podem ser adotadas com o uso da Inteligência Artificial, sobretudo a Descoberta de Conhecimento em Bases de Dados. Este trabalho propõe-se a criação de uma metodologia e aplicação que realize a Mineração de Dados com informações textuais vinculados a dados geolocalizados em uma Rede Social, com o intuito de promover a Recomendação Social. Entretanto, as abordagens na construção dos Sistemas de Recomendação apresentam algumas deficiências na filtragem dos resultados e na forma que estes são sugeridos aos usuários. A pesquisa busca a solução destas deficiências e aborda temas que ainda carecem de pesquisas mais efetivas e resultados consolidados.

Le but de cette thèse est de comprendre comment les entreprises peuvent faire augmenter une expérience donnant naissance à l’engagement des consommateurs grâce aux nouveaux médias (comme les vidéos du Web, les dispositifs de téléphonie mobile et la télévision "traditionnelle") afin de stimuler le comportement actif des clients et de redéfinir des stratégies commerciales de marketing. Nous avons structuré notre analyse sur trois études d’approche.Dans la première étude, nous avons décrit comment l'engagement personnel avec le contenu et l'engagement social interactif (résultant du sens perçu de la communauté, du sentiment d'appréciation intrinsèque et de la participation à l'expérience) influencent différemment le comportement actif et passif sur des sites de télévision sociale. Nous avons testé des hypothèses en estimant un modèle d'équation structurale avec les données d'une enquête sur un groupe de 814 utilisateurs de télévision sociale aux Etats-Unis et en Europe. Dans la deuxième étude, nous examinons l'influence de l'intrusion dans la vie privée sur la relation entre l'engagement expérientiel (c'est à dire l'engagement personnel et l'engagement interactif et social) et l'utilisation active et passif et nous avons testé ces hypothèses (379 utilisateurs) en tenant compte de services de géolocalisation sur téléphonie mobile. Dans la troisième étude, nous avons élargi notre cadre conceptuel et étudié les effets de l'engagement social interactif sur l'identité sociale et l'appréciation des marques. Le modèle a été validé expérimentalement en menant une enquête sur des pages de fans de Facebook de 20 grandes marques internationales situées en Europe et aux Etats-Unis (panel de 387 personnes). Les résultats émergeant des trois études prouvent que l'engagement expérientiel a des effets positifs sur le comportement du consommateur (actif et passif) en ligne et qu'il contraste avec l'effet négatif de l'atteinte à la vie privée. Les résultats obtenus confirment les effets positifs de l'engagement social et interactif sur les rapports affectifs des consommateurs pour une marque et le plein effet de l'identité sociale. De manière plus spécifique, les annonceurs publicitaires, qui forcent les expériences pouvant influencer l'engagement social et interactif, peuvent aussi influer sur l'identité sociale et le rapport avec une marque. / The thesis aims to understand how companies can leverage on consumer experiential engagement in new-media based social media environments (using video on the web, handheld devices and web 2.0) in order to stimulate active behavior and redefine commercial marketing strategies. We structure our analysis on a three studies approach. The first study describes how Personal Engagement with the content and Social-Interactive Engagement (resulting from the perceived sense of community, intrinsic enjoyment and participation experience) differently influence both active and passive behavior. We test hypotheses with survey data from a sample of 814 US and EU social TV users. In study 2 we examine the influence of privacy intrusiveness on the relation between Experiential Engagement (Personal and Social-interactive Engagement) and active and passive use and we test it (n=379) with reference to mobile location-based social networking applications in EU and US. In study 3 we develop a conceptual model in which social-interactive engagement influences social identity directly and brand love indirectly through the mediating effect of social identity. The model was empirically validated (n=387) on the Facebook fan pages of 20 leading international brands in EU and the US. Findings emerging from the three studies show that Experiential Engagement has positive effects on the consumer behavior online (active and passive) and it may contrast the negative effect of privacy intrusiveness. The results obtained show also a positive effects of social-interactive engagement on consumer-brand affective relationships (brand love) and the full mediating effect of social identity. More specifically advertisers, leveraging on experiences that influence social-interactive engagement can influence the social identity and the relationship with the brand