Centre de documentation
scientifique

Using attributes explicitly reflecting user preference in a self-attention network for next POI recommendation / Ruijing Li in ISPRS International journal of geo-information, vol 11 n° 8 (August 2022)  

Public [article]  inISPRS International journal of geo-information > vol 11 n° 8 (August 2022) . - n° 440 Titre : Using attributes explicitly reflecting user preference in a self-attention network for next POI recommendation Type de document : Article/Communication Auteurs : Ruijing Li, Auteur ; Jianzhong Guo, Auteur ; Chun Liu, Auteur ; et al., Auteur Année de publication : 2022 Article en page(s) : n° 440 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Analyse spatiale [Termes IGN] apprentissage profond [Termes IGN] attention (apprentissage automatique) [Termes IGN] distance [Termes IGN] filtrage d'information [Termes IGN] New York (Etats-Unis ; ville) [Termes IGN] point d'intérêt [Termes IGN] réseau social géodépendant [Termes IGN] Tokyo (Japon) Résumé : (auteur) With the popularity of location-based social networks such as Weibo and Twitter, there are many records of points of interest (POIs) showing when and where people have visited certain locations. From these records, next POI recommendation suggests the next POI that a target user might want to visit based on their check-in history and current spatio-temporal context. Current next POI recommendation methods mainly apply different deep learning models to capture user preferences by learning the nonlinear relations between POIs and user preference and pay little attention to mining or using the information that explicitly reflects user preference. In contrast, this paper proposes to utilize data that explicitly reflect user preference and include these data in a deep learning-based process to better capture user preference. Based on the self-attention network, this paper utilizes the attributes of the month of the check-ins and the categories of check-ins during this time, which indicate the periodicity of the user’s work and life and can reflect the habits of users. Moreover, considering that distance has a significant impact on a user’s decision of whether to visit a POI, we used a filter to remove candidate POIs that were more than a certain distance away when recommending the next POIs. We use check-in data from New York City (NYC) and Tokyo (TKY) as datasets, and experiments show that these improvements improve the recommended performance of the next POI. Compared with the state-of-the-art methods, the proposed method improved the recall rate by 7.32% on average. Numéro de notice : A2022-647 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.3390/ijgi11080440 Date de publication en ligne : 04/08/2022 En ligne : https://doi.org/10.3390/ijgi11080440 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=101463 [article]

A GIS representation framework for location-based social media activities / Xuebin Wei in Transactions in GIS, vol 26 n° 3 (May 2022)  

Public [article]  inTransactions in GIS > vol 26 n° 3 (May 2022) . - pp 1444 - 1464 Titre : A GIS representation framework for location-based social media activities Type de document : Article/Communication Auteurs : Xuebin Wei, Auteur ; Xiaobai Yao, Auteur Année de publication : 2022 Article en page(s) : pp 1444 - 1464 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Géomatique web [Termes IGN] cadre conceptuel [Termes IGN] données issues des réseaux sociaux [Termes IGN] environnement géographique virtuel [Termes IGN] Facebook [Termes IGN] modèle conceptuel de données spatio-temporelles [Termes IGN] ontologie [Termes IGN] relations sociales [Termes IGN] représentation des données [Termes IGN] réseau social géodépendant [Termes IGN] système d'information géographique [Termes IGN] Time-geography Résumé : (auteur) The past couple of decades have witnessed tremendous growth of location-based social media activities (LBSMA) data in virtual spaces, including virtual geographic environments. Such data become innovative resources for the analysis of human activities. Meanwhile, a shift of human interactions from geographical spaces to virtual spaces has been observed. Although this is an exciting research opportunity, it also imposes significant challenges on GIScience, as current GIS representation models are no longer sufficient to handle the increased sophistication of human activities data. This research formalizes an ontology for LBSMA data and a conceptual framework for representing such data in GIS. The framework contributes to GIScience as it enables interconnections of human activities in both the physical and virtual worlds to be represented, organized, retrieved, analyzed, and visualized. The proposed GIS representation model integrates a social dimension into the existing spatial–temporal representation models and allows data analysis in the spatial–temporal–social (STS) dimensions. The research tested this conceptual framework with a prototype and a case study using Facebook data. The prototype and the case study prove that the proposed framework can significantly enhance GIS capabilities for data organization, retrieval, and analysis of LBSMA data in STS dimensions. Numéro de notice : A2022-477 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE/SOCIETE NUMERIQUE Nature : Article DOI : 10.1111/tgis.12929 Date de publication en ligne : 02/05/2022 En ligne : https://doi.org/10.1111/tgis.12929 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100825 [article]

Modular multi-dimensional tool for emergency evacuation including location-based social network data / Ilil Blum Shem-Tov in Journal of location-based services, vol 16 n° 1 (March 2022)  

Public [article]  inJournal of location-based services > vol 16 n° 1 (March 2022) . - pp 54 - 75 Titre : Modular multi-dimensional tool for emergency evacuation including location-based social network data Type de document : Article/Communication Auteurs : Ilil Blum Shem-Tov, Auteur ; Shlomo Bekhor, Auteur Année de publication : 2022 Article en page(s) : pp 54 - 75 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Géomatique web [Termes IGN] distribution spatiale [Termes IGN] données issues des réseaux sociaux [Termes IGN] données localisées des bénévoles [Termes IGN] origine - destination [Termes IGN] réseau social géodépendant [Termes IGN] secours d'urgence [Termes IGN] téléphone intelligent Résumé : (auteur) This paper presents the concept of a modular multi-dimensional tool (MMDT) for evacuation planning models. The goal of MMDT is to propose alternative route and destination locations that can be evaluated and compared to one another. The proposed tool can represent a very large number of scenarios and its strength is in its modularity and efficiency. The MMDT can be applied using both conventional evacuation models and decentralised personalised evacuation models based on Location-Based Social Networks (LBSN) to reduce overall evacuation times. Large-scale test cases using anonymous LBSN data illustrate the MMDT on several scenarios. Results indicate a significant reduction in evacuation times when using decentralised personal evacuation. Numéro de notice : A2022-389 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Article DOI : 10.1080/17489725.2021.1990422 Date de publication en ligne : 16/11/2021 En ligne : https://doi.org/10.1080/17489725.2021.1990422 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100679 [article]

Contextual location recommendation for location-based social networks by learning user intentions and contextual triggers / Seyyed Mohammadreza Rahimi in Geoinformatica, vol 26 n° 1 (January 2022)  

Public [article]  inGeoinformatica > vol 26 n° 1 (January 2022) . - pp 1 - 28 Titre : Contextual location recommendation for location-based social networks by learning user intentions and contextual triggers Type de document : Article/Communication Auteurs : Seyyed Mohammadreza Rahimi, Auteur ; Behrouz Far, Auteur ; Xin Wang, Auteur Année de publication : 2022 Article en page(s) : pp 1 - 28 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Géomatique web [Termes IGN] analyse spatiale [Termes IGN] comportement [Termes IGN] contenu généré par les utilisateurs [Termes IGN] covariance [Termes IGN] données spatiotemporelles [Termes IGN] historique des données [Termes IGN] interface web [Termes IGN] mobilité territoriale [Termes IGN] prise en compte du contexte [Termes IGN] réseau social géodépendant [Termes IGN] service fondé sur la position [Termes IGN] système de recommandation Résumé : (auteur) Location recommendation methods suggest unvisited locations to their users. Many existing location recommendation methods focus on the spatial, social and temporal aspects of human movements. However, contextual information is also invaluable to location recommendation methods and has the great potential for explaining what triggers users to show different behaviors. CLR learns the response of the users to contextual variables based on their own history and the history of similar behaving users. In this paper, we propose a contextual location recommendation method named Contextual Location Recommendation (CLR) that learns the intention and spatial responses of users to various contextual triggers using the historical check-in and contextual information. CLR starts with a co-variance analysis to reduce dimensionality of the check-in data and then uses an optimized version of the random walk with restart to extract hidden user responses to contextual triggers. A tensor factorization is used to build a latent-factor model to predict the user’s intention response with the given set of contextual triggers. Based on the intention response of the user, a contextual spatial component identifies a set of matching locations accessible to the user by estimating the probability distribution of the location of the user and the popularity probability of locations under the contextual settings. Experimental results on three real-world datasets show that CLR improves the recommendation precision by 35% compared to the best-performing baseline recommendation method. Numéro de notice : A2022-203 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Article DOI : 10.1007/s10707-021-00437-y Date de publication en ligne : 02/06/2021 En ligne : https://doi.org/10.1007/s10707-021-00437-y Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100008 [article]

Domain adaptation for urban scene segmentation / Antoine Saporta (2022)  

Public Titre : Domain adaptation for urban scene segmentation Type de document : Thèse/HDR Auteurs : Antoine Saporta, Auteur ; Matthieu Cord, Directeur de thèse Editeur : Paris : Sorbonne Université Année de publication : 2022 Importance : 147 p. Format : 21 x 30 cm Note générale : bibliographie Thèse de Doctorat de Sorbonne Université, spécialité Informatique Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] apprentissage profond [Termes IGN] classification non dirigée [Termes IGN] entropie [Termes IGN] Mapillary [Termes IGN] navigation autonome [Termes IGN] réseau antagoniste génératif [Termes IGN] scène urbaine [Termes IGN] segmentation sémantique [Termes IGN] vision par ordinateur Index. décimale : THESE Thèses et HDR Résumé : (auteur) This thesis tackles some of the scientific locks of perception systems based on neural networks for autonomous vehicles. This dissertation discusses domain adaptation, a class of tools aiming at minimizing the need for labeled data. Domain adaptation allows generalization to so-called target data that share structures with the labeled so-called source data allowing supervision but nevertheless following a different statistical distribution. First, we study the introduction of privileged information in the source data, for instance, depth labels. The proposed strategy, BerMuDA, bases its domain adaptation on a multimodal representation obtained by bilinear fusion, modeling complex interactions between segmentation and depth. Next, we examine self-supervised learning strategies in domain adaptation, relying on selecting predictions on the unlabeled target data, serving as pseudo-labels. We propose two new selection criteria: first, an entropic criterion with ESL; then, with ConDA, using an estimate of the true class probability. Finally, the extension of adaptation scenarios to several target domains as well as in a continual learning framework is proposed. Two approaches are presented to extend traditional adversarial methods to multi-target domain adaptation: Multi-Dis. and MTKT. In a continual learning setting for which the target domains are discovered sequentially and without rehearsal, the proposed CTKT approach adapts MTKT to this new problem to tackle catastrophic forgetting. Note de contenu : 1- Introduction 2- Unsupervised domain adaptation 3- Leveraging priviledge information for unsupervised domain adaptation 4- Estimating and exploiting confident pseudo-labels for self-training 5- Adaptation to multiple domains 6- Conclusion Numéro de notice : 24079 Affiliation des auteurs : non IGN Thématique : IMAGERIE/INFORMATIQUE Nature : Thèse française Note de thèse : Thèse de Doctorat : Informatique : Sorbonne Université : 2022 Organisme de stage : Institut des Systèmes Intelligents et de Robotique DOI : sans En ligne : https://theses.hal.science/tel-03886201 Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=102213

Urban infrastructure audit: an effective protocol to digitize signalized intersections by mining street view images / Xiao Li in Cartography and Geographic Information Science, vol 49 n° 1 (January 2022)  

Permalink

Connecting family trees to construct a population-scale and longitudinal geo-social network for the U.S. / Caglar Koylu in International journal of geographical information science IJGIS, vol 35 n° 12 (December 2021)  

Permalink

A topology-based graph data model for indoor spatial-social networking / Mahdi Rahimi in International journal of geographical information science IJGIS, vol 35 n° 12 (December 2021)  

Permalink

The geography of social media data in urban areas: Representativeness and complementarity / Alvaro Bernabeu-Bautista in ISPRS International journal of geo-information, vol 10 n° 11 (November 2021)  

Permalink

Towards generating network of bikeways from Mapillary data / Xuan Ding in Computers, Environment and Urban Systems, vol 88 (July 2021)  

Permalink

Constructing and analyzing spatial-social networks from location-based social media data / Xuebin Wei in Cartography and Geographic Information Science, vol 48 n° 3 (May 2021)  

Permalink

Joint promotion partner recommendation systems using data from location-based social networks / Yi-Chung Chen in ISPRS International journal of geo-information, vol 10 n° 2 (February 2021)  

Permalink

Incorporating memory-based preferences and point-of-interest stickiness into recommendations in location-based social networks / Hang Zhang in ISPRS International journal of geo-information, vol 10 n° 1 (January 2021)  

Permalink

Introducing diversion graph for real-time spatial data analysis with location based social networks / Sameera Kannangara (2021)  

Permalink

RegNet: a neural network model for predicting regional desirability with VGI data / Wenzhong Shi in International journal of geographical information science IJGIS, vol 35 n° 1 (January 2021)  

Permalink