Fig 1 - uploaded by Nuno C. Marques
Content may be subject to copyright.
Source publication
In this paper, we propose a framework named UMT (User-profile Modeling based on Transactional data) for modeling user group profiles based on the transactional data. UMT is a generic framework for a pplication systems that keep the historical transactions of their users. In UMT, user group profiles consist of three types: basic information attribut...
Contexts in source publication
Context 1
... user group profile UGP=<ugp_id, {(attr 1 , val 1 ), (attr 2 , val 2 ),…}> is a 2-tuple with a key and a set of (attribute, value)- pairs. Figure 1 shows the architecture of UMT. UMT consists of three modules: the data preparing module, the user modeling module, and the recommendation generating module. ...
Similar publications
In a typical Monte Carlo simulation (MCS), it is required to assign probability distribution functions (PDFs) for uncertain parameters. Due to lack of or irrelevance of historical whole-life costing (WLC) data, these functions are usually assumed. The research work that underpins this paper aimed to identify the significance of errors in various in...
In this paper, we derived a probability distribution model that considers joint probability of discrete and continuous distribution models to estimate extreme values of ocean environment factor. We aim to improve the design standard research in ocean engineering projects. We use this new model to calculate the joint extreme value of multiple ocean...
Citations
... El primer artículo encontrado en esta investigación trata sobre un sistema llamado COOL-TOUR donde los amantes al turismo de montaña formaron una comunidad donde podían ingresar datos de recorridos y otros usuarios podían utilizar esa información para mejorar su experiencia (Blanzieri, & Ebranati, 2000). Varios de los artículos publicados en este período de tiempo, explican como la inteligencia artifial brinda información personalizada a partir de los intereses, preferencias (Fink, J., y Kobsa, A., 2002) y de datos transaccionales de los usuarios (Yang, Y., y Marques, N. C., 2005) obtenidos en su mayoria de sitios web de hoteles (Law, 2005), aerolíneas y trenes (Camacho, D., Aler, R., Borrajo, D., y Molina, J. M., 2006) por medio del uso de algoritmos (Goh, C., y Law, R., 2003) que establecen modelos predictivos muy precisos sobre la demanda de viajes desde distintos países (Cho, 2003) o patrones de llegada de turistas a ciertos sitios (Hadavandi, E., Ghanbari, A., Shahanaghi, K., y Abbasian-Naghneh, S., 2011). ...
... El primer artículo encontrado en esta investigación trata sobre un sistema llamado COOL-TOUR donde los amantes al turismo de montaña formaron una comunidad donde podían ingresar datos de recorridos y otros usuarios podían utilizar esa información para mejorar su experiencia (Blanzieri, & Ebranati, 2000). Varios de los artículos publicados en este período de tiempo, explican como la inteligencia artifial brinda información personalizada a partir de los intereses, preferencias (Fink, J., y Kobsa, A., 2002) y de datos transaccionales de los usuarios (Yang, Y., y Marques, N. C., 2005) obtenidos en su mayoria de sitios web de hoteles (Law, 2005), aerolíneas y trenes (Camacho, D., Aler, R., Borrajo, D., y Molina, J. M., 2006) por medio del uso de algoritmos (Goh, C., y Law, R., 2003) que establecen modelos predictivos muy precisos sobre la demanda de viajes desde distintos países (Cho, 2003) o patrones de llegada de turistas a ciertos sitios (Hadavandi, E., Ghanbari, A., Shahanaghi, K., y Abbasian-Naghneh, S., 2011). ...
... The new method uses social relations between people as a rich information resource. In most recommender systems, information source includes user purchases transactions, user ratings to items, demographic information of user, and item profiles ( García-Crespo et al., 2009 ;Huang & Bian, 2009 ;Pi, Ji & Yang, 2018 ;Schiaffino & Amandi, 2009 ;Yang & Marques, 2005 ). Whereas based on Homophily Principle ( McPherson, Smith-Lovin & Cook, 2001 ) in Social Networks, similarity causes communication and relationship. ...
... Tourism recommender systems were almost based on collaborating filtering approach ( García-Crespo et al., 2009 ;Shelar, Kamat, Varpe & Birajdar, 2018 ;Yang & Marques, 2005 ) and hybrid approaches ( Huang & Bian, 2009 ;Schiaffino & Amandi, 2009 ). Also, most of them used both explicit and implicit methods to identify user interest ( García-Crespo et al., 2009 ;Huang & Bian, 2009 ). ...
... They used location, time and weather as criteria. Yang and Marques (2005) offer a tourism recommendation based on collaborative filtering approach and implicit criteria. ...
Web 2.0 and its services, such as social networks, have significantly influenced various businesses, including e-commerce. As a result, we face a new generation of e-commerce called Social Commerce. On the other hand, in the tourism industry, a variety of services and products are provided. The dramatic rise in the number of options in travel packages, hotels, tourist attractions, etc. put users in a difficult situation to find what they need. For a reason, tourism recommender systems have been considered by researchers and businesses as a solution. Since tourist attractions are often the reason for travelling, this research proposes a social-hybrid recommender system in the context of social commerce that recommends tourist attractions. The purpose of the research is presenting a personalized list of tourist attractions for each tourist based on the similarity of users' desires and interests, trust, reputation, relationships, and social communities. Compared with the traditional methods, collaborative filtering, content-based, and hybrid, the advantage of the proposed method is the use of various factors and the inclusion of trust factors in recommendation resources, (such as outlier detection in user ratings), and employing social relationships among individuals. The experimental results show the superiority of the proposed method over other common methods. The proposed method can also be used to recommend other products and services in the tourism industry and other social commerce.
Read more: https://authors.elsevier.com/c/1abeE3PiGTFIFF
... This has been mainly used in outdoor tours for a path or sight recommendations. Indeed, mobile devices have been used for finding out information about the cultural sights of a city or even a whole country and have been used by Speta (García- Crespo et al. 2009) and Speta II (García- Crespo et al. 2010), PinPoint (Roth 2002), m-ToGuide prototype (Kamar 2003) and UMT (Yang & Marques, 2005). ...
Personalized software has been extensively used in museum guides and recommendation systems for tours because it provides added value to the interaction of the user with cultural heritage. However, this added value can only be confirmed through an evaluation experiment. Therefore, this chapter presents the indoor and outdoor personalized tours for exploring cultural heritage. More specifically, the chapter focuses on the methods used for personalization, the technology employed to improve interaction, and the evaluation experiments that have been implemented. The evaluation experiments are further categorized with respect to the method of evaluation and the criteria used. Finally, it provides a discussion on the main conclusions drawn from the researchers conducting evaluation experiments of the system performing indoor and outdoor tours for the exploration of cultural heritage.
... In micro-blog social networks, most recommender systems adopt collaborative filtering or hybrid strategies to provide users with numerous valuable information from similar users [16][17][18]. However, existing works based on similar users often supply repeated items, which lack interest diversity. ...
... As a result, it is time consuming to find interest from all similar followee users. Diverse interest mining from representative users can be used to decision analysis, vote, e-commerce, web search engines, socialized recommender systems and museum visitors' quality of experience and other fields [4,16,[18][19][20][21]. Thus, it is reasonable that the followees could be replaced by a dissimilar and representative group to GUP for recommendation [7,10,19]. ...
... Based on user behavior modeling and user satisfaction capturing, Tsiropoulou et al. [27] adopted optimal and heuristic strategies to design a social recommendation and personalization approach, which can recommend a set of exhibits to the visitors. Considering basic information attributes, synthetic attributes and probability distribution attributes, Yang et al. [16] designed a framework named UMT (User-profile Modeling based on Transactional data) for modeling group profiles, which is suitable for personalization of transaction-based commercial application systems. Using social annotations, Mezghani et al. [28] contributed on characteristics of social user and techniques about modeling tag-based user profile and updating corresponding profiles for recommendation. ...
How to find a user’s interest from similar users a fundamental research problems in socialized recommender systems. Despite significant advances, there exists diversity loss for the majority of recommender systems. With this paper, for expanding the user’s interest, we overcome this challenge by using representative and diverse similar users from followees. First, we model a personal user profile vector via word2vec and term frequency mechanisms. According to user profiles and their follow relationships, we compute content interaction similarity and follow interaction similarity. Second, by combining two kinds of interaction similarity, we calculate the social similarity and discover a diverse group with coverage and dissimilarity. The users in a diverse group can distinguish each other and cover the whole followees, which can model a group user profile (GUP). Then, by tracking the changes of followee set, we heuristically adjust the number of diverse group users and construct an adaptive GUP. Finally, we conduct experiments on Sina Weibo datasets for recommendation, and the experimental results demonstrate that the proposed GUP outperforms conventional approaches for diverse recommendation.
... Furthermore, a user's characteristics may change over time. Some systems that use stereotypical techniques for personalizing the presentation of cultural sights to tourists are AVANTI (Fink et al. 1996, Fink et al. 1998), INTRIGUE (Ardissono et al. 2003) and UMT (Yang and Marques 2005). However, a main problem that such systems encounter is that each user differentiates from all the others in many ways. ...
This paper makes a review on personalisation systems specialised for cultural tourism. Tourists interested in cultural heritage have different requirements from tourist recommendation systems than other users. Therefore, emphasis is given on recommendation systems for city tours and museum guides. More specifically, systems for PC, PDAs and mobile phones are discussed as well as the methods and the technologies used.
... Traveller, for instance, uses the following the characteristics: category, place orTable 1 Categorization of recommender systems based on the features they recommend. Hotel Flight Restaurants- café-shops Attractions – sights General information Map guidance Packages Traveller (Schi- affino and Amandi, 2009) Travel Planner (Chin and Porage, 2001) AVANTI (Fink et al., 1996Fink et al., , 1998) MAIS Project (Corallo et al., 2005) Speta (GarcíaCerspo et al., 2009) m-ToGuide prototype (Kamar, 2003) ITR (Ricci et al., 2003) MastroCARonte (Console et al., 2003) ATA (Linden et al., 1997) m-ToGuide prototype (Kamar, 2003) MAIS Project (Corallo et al., 2005) Travel Planner (Chin and Porage, 2001) Traveller (Schi- affino and Amandi, 2009) PTA (Coyle and Cunningham, 2003) Entreé (Burke, 2000) CAPA (Tung and Soo, 2004) MastroCARonte (Console et al., 2003) CRUMPET (Posland et al., 2001) CATIS (Pashtan et al., 2003) Traveller (Schi- affino and Amandi, 2009) Travel Planner (Chin and Porage, 2001) Speta (GarcíaCerspo et al., 2009)Console et al., 2003) CRUMPET (Posland et al., 2001) Traveller (Schiaffino and Amandi, 2009) Travel Planner (Chin and Porage, 2001) AVANTI (Fink et al., 1996Fink et al., , 1998) INTRIGUE (Ardissono et al., 2003) Hinze and Voisard (2003) Speta (Cerspo et al., 2009) Gulliver's Genie (Hristova et al., 2003; O'Grady and O'Hare, 2004) MobiDENK (Krösche et al., 2004; Baldzer et al., 2004) PinPoint (Roth, 2002) m-ToGuide prototype (Kamar, 2003) UMT (Yang and Marques, 2005) ITR (Ricci et al., 2003) Cyberguide (Abowd et al., 1997) GUIDE system (Cheverst et al., 2002) WebGuide (WebGuide, 2001; Fink and Kobsa 2002) Sightseeing4U (Scherp and Boll, 2004) LoL@ (Anegg et al., 2002) Speta (García-Cerspo et al., 2009) PinPoint (Roth, 2002) m-ToGuide prototype (Kamar, 2003)Table 2 Criteria used for flight recommendation. ATA (Linden et al., 1997) m-ToGuide prototype (Kamar, 2003) Travel Planner (Chin and Porage, 2001) Traveller (Schiaffino and Amandi, 2009) PTA (Coyle and Cunningham, 2003) Historical or artistic value p destination, price, means of transport, transport company (e.g. ...
... Pouloudi et al. (2002) summarized the Internet users' profile into seven e-social factors, namely: region/geography, culture, legal/regulation/policy, economic,Table 8 Categorization of tourism recommendation systems with respect to the dimension about group or individual user modeling. Stereotypes/group modeling Individual UM Combination Entreé (Burke, 2000) Travel Planner (Chin and Porage, 2001) AVANTI (Fink et al., 1996Fink et al., , 1998) INTRIGUE (Ardissono et al., 2003) Maw and Naing (2006) UMT (Yang and Marques, 2005) Soe et al. ...
... The system then matches the user preferences with items in a catalog of destinations (described in the same language).Table 10 Categorization of tourism recommendation systems with respect to the dimension about the user model's duration. Long-term Short-term PTA (Coyle and Cunningham, 2003) CAPA (Tung and Soo, 2004) Huang and Bian (2009) GUIDE system (Cheverst et al., 2000) WebGuide (WebGuide, 2001; Fink and Kobsa, 2002), Sightseeing4U (Scherp and Boll, 2004) IMA (Gervais et al., 2007) CATIS (Pashtan et al., 2003) Travel Planner (Chin and Porage, 2001) AVANTI (Fink et al., 1996Fink et al., , 1998) PTS (Srivihok and Sukonmanee, 2005) Gulliver's Genie (Hristova et al., 2003; O'Grady and O'Hare, 2004) PinPoint (Roth, 2002) m-ToGuide prototype (Kamar, 2003) Niaraki and Kim (2009) ITR (Ricci et al., 2003) Entreé (Burke, 2000)Tung and Soo, 2004) Cyberguide (Abowd et al., 1997) GUIDE system (Cheverst et al., 2000) WebGuide (WebGuide, 2001; Fink and Kobsa, 2002) Sightseeing4U (Scherp and Boll, 2004) MastroCARonte (Console et al., 2003) CATIS (Pashtan et al., 2003) MAIS Project (Corallo et al., 2005) INTRIGUE (Ardissono et al., 2003) Gulliver's Genie (Hristova et al., 2003; O'Grady and O'Hare, 2004) MobiDENK (Krösche et al., 2004; Baldzer et al., 2004) PinPoint (Roth, 2002) m-ToGuide prototype (Kamar, 2003) Niaraki and Kim (2009) PRSET (Srisuwan and Srivihok, 2008) Iwasaki et al. (2007) ATA (Linden et al., 1997) ITR (Ricci et al., 2003) Speta (García-Cerspo et al., 2009) Maw and Naing (2006) UMT (Yang and Marques, 2005) Soe et al. (2006) PTA (Coyle and Cunningham, 2003) Entreé (Burke, 2000) Huang and Bian (2009) Traveller (Schiaffino and Amandi, 2009 Travel Planner (Chin and Porage, 2001) PTS (Srivihok and Sukonmanee, 2005) Hinze and Voisard (2003) For this purpose, different methods have been proposed in the literature. For example, some approaches may use simple syntactic comparisons among the attributes of the considered items or services and the user profiles' attributes whereas other methods have predictive capabilities in order to improve the system's performance. ...
Internet has significantly influenced the tourism sector providing a great variety of services and products online. However, the number of choices has increased so dramatically that is very difficult for the consumers to find what they are looking for. For this purpose, recommendation systems for tourism have attracted a lot of research energy and interest. The main characteristic of these systems is that they can personalize their recommendations to each user interacting with the system. Personalization is even more essential for tourism recommendation systems used in handheld devices where the screen is even smaller and the presentation capabilities are limited. This paper addresses these problems and provides some development steps for a tourism recommendation system by making a state of the art in personalized e-tourism services both in computers and handheld devices as well as a review of the user modeling and personalization techniques used in these systems. Furthermore, the theories used for the improvement of the personalization procedure in tourism recommendation systems; their applications and evaluation are discussed.
... ity between user and tag. K. R. Bayyapu and P. Dolog in [2] tries to solve the problems of sparse data and low quality of tags from related domains. They suggest using tag neighbors for tag expression expansion. However the tag neighbors are based on the content of documents. We propose another approach to extend the tag set by the group profiling. [19] uses a framework of User-Profile Modeling based on Transactional data for modeling user group profiles based on the transactional data which can incorporate external information, either by means of an internal knowledge base or on dynamic data supplied by a specific information extraction system. Such user group profiles consist of thre ...
In this paper, we propose a spectral cluster-ing approach for users and documents group modeling in order to capture the common preference and relatedness of users and doc-uments, and to reduce the time complexity of similarity calculations. In experiments, we investigate the selection of the optimal amount of clusters. We also show a reduc-tion of the time consuming in calculating the similarity for the recommender systems by se-lecting a centroid first, and then compare the inside item on behalf of each group.
Currently, tourism providers build tourist models by collecting some specific pieces of information and then combining the knowledge they have about the groups to which current tourists belong. This paper presents the BaliaTour user modelling and Recommender System, which combines several techniques and methodologies in order to enhance the modelling process when scarce of data about an individual tourist is available. The core of the model is based on the predictions made over stereotypes as a initial characterization of the user profile. The modelling is further refined and enhanced by the combination of explicit preferences and ratings provided by the user. As a result, the proposed approach takes advantage of every information piece known about tourists in tourism ecosystems. The main advantage of BaliaTour is to minimize the main drawbacks of each of the existing user modelling techniques to obtain a user model.
Personalization has an important role in information systems. It is an effective solution for reducing complexity when searching information. In this way, the user feels like the system was developed for him/her. In this context, personalization can be seen as an optimization problem. To this end, we propose a multi-criteria decision making approach to personalize system. The proposed approach has been validated by applying it to personalize a system in intelligent transport field.
