constrained device in pervasive environment, we consider a deployment strategy of services ..... Taldea consist of an android application. It is composed of ... and pink is higher in the botanical ontology because the pink color is frequently ...
Design and development of semantic application for communities Ghada Ben Nejma, Philippe Roose, Dalmau Marc
Jérôme Gensel, Ghorbali Mohamed Amine
LIUPPA Laboratory T2I Research team 2 Allée du Parc de Montaury 64600 Anglet, France {gbennej, roose, dalmau}@iutbayonne.univ-pau.fr
LIG Laboratory / STEAMER Research team 681 rue de la Passerelle, BP72 38402 Saint Martin d’Hères cedex, France {Jerome.Gensel, mohamed-amine.ghorbali}@imag.fr
Abstract—In the recent years, social applications have competed to incorporate emerging technologies and innovative aspects like semantics technologies, location based services, data ephemerality to ensure user privacy, etc. Although few applications combine these aspects, our challenge is to gather them in applications for communities. Likewise due to the limited resources of mobile devices and the growing number of services, context awareness becomes a major requirement in social computing. Applications must be adapted to context constraints (such as the host’s resources, the services properties and the surrounding environments) and must provide appropriate or personalized services. This paper presents features and architectural principles upon which is based the design and the development of our application for fostering spontaneous communities. This kind of communities answers the needs that cannot be solved in perennial communities or may be more effective in circumstantial situations (e.g. accidents, natural disaster, crisis, fire, etc.). The application provides multiple services in order to satisfy the needs of users sharing a common interest or a common purpose. As a solution for resourceconstrained device in pervasive environment, we consider a deployment strategy of services based on a semantic web services matching algorithm. The proposed approach has been validated through a prototype called Taldea used by the visitors of a botanical garden. Keywords—Communities applications, Semantic technologies, Location based services, Temporary Social network, User Policies rules, Dynamic deployment of services.
I.
INTRODUCTION
The popularity of social networks and online communities has grown exponentially. They are more and more used to connect, communicate, and share information with others. Several researches [1] [2] outline the limitations of centralized social networks such as privacy threatens, no ownership of data, no interoperability, less trust, no support for collective knowledge, etc. To overcome the limitations of existing centralized social networks, distributed social networks are proposed based on a P2P architecture that allows data download from end users (e.g. LifeSocial [3], PeerSoN [4]). Also, the native social mobile (unlike the mobile social network client) starts to disrupt centralized social network [5, 6]. These are social networks but with a special focus on mobility by putting forward mobile communication, locationbased services, and augmented reality, etc.
The democratization of the mobile devices (e.g. smartphones, touchpads, laptops, sensors, etc.) has made information accessible to anyone at anytime and from anywhere and facilitates the capture of physical context. Pervasive computing offers great abilities for developing applications that are flexible, adaptable, and capable of acting autonomously on behalf of users. The intersection of the two domains (i.e. social networks and pervasive computing) gives rise to emerging fields of research called Pervasive Social Computing [7]. Bringing context-awareness into social networks increases knowledge about user’s situation and social interaction in a virtual or physical context. The transparent architecture of such applications enables user to control their data. This advantage gives social mobile applications a significant edge over centralized social network. Systems in the area of mobile and pervasive social computing refer more and more to the notion of community: ‘in society, individuals very often communicate and interact with others in groups. When groups of users have common interests or undertake common activities, they share, on a regular basis, information with each other. Therefore universal and pervasive services and applications must also enable users to reach and interact with groups consisting of many individuals’ [8]. This granularity of people (i.e. community) is considered in our application as distinct social entities that should be supported with services as single user. This paper describes the development of social software, called Taldea that helps user to access to communities and organize exchanges with other members. Our work focuses more particularly on spontaneous, short-lived communities. Intuitively, it is the type of community that best matches with circumstantial, accidental, incidental or fortuitous situations. We define this new type of communities as ‘a spontaneous group of individuals having a common interest or purpose related to a circumstantial situation and relative to a geographical territory’. This shape of community can meet specific needs which are not taken into account by perennial communities. For example, imagine a person walking in the park and who finds a wounded animal. She takes her Smartphone and creates a community to ask for help from people around. She can share the location, and some photos of the animal. In return, some people can advise her by texting or move and join her on the
scene through the sharing of location guided by location based services.
believe that application would ensure a user friendly way to use these technologies.
One of our challenges is to build our semantic application Taldea on a set of standards that are developed by the W3C in order to describe formally a community. These standards make it possible for software agents to understand exchanged information without the ambiguity, complex processing, and rigidity brought by other representation formalisms (e.g. natural language, relational database). As a basis for this initiative, we aim at using the result of our previous work [9], an ontology that formally describes a community using multiple standards such as FOAF1, SIOC2, SKOS3, Owl-Time4 and GeoRSS5, OWL-S6. The model is structured over a set of abstract entities, each describing physical or conceptual objects including Interest, Member, Lifespan, Resource, Location, Type. This set of concepts is shared by each community.
B. Temporary social network In the case of known social networks, the shared content is archived, recorded and, potentially, monitored. With these troublesome issues, the concept of temporary social network appears. TSN is a social network where the content is selfdestructed which enhances to the privacy issues. There is no record, no link or opportunity to share it again. The content sent disappears once some rules are checked. Among the applications that follow this trend:
In addition, this work benefits from the use of our Kalimucho [10] platform in order to dynamically deploy components assembled to compose the application. The paper is organized as follows. Section 2 gives an overview of some related work. The architecture and the deployment strategy of Taldea services are described in section 3. Then, the application Taldea that validates our approach is described in section 4. After presenting the components of the architecture, section 5 analyzes the experimental results. Finally, a conclusion summarizes the presented approach and outlines some future work. II.
RELATED WORK
In this section we present some research in the area of social computing related to the four features of our application: Community application, Semantic technologies, location based services, ephemerality in social media. Our system is not the first to incorporate one of these dimensions but it can be seen as one of the few attempts to combine the four. A. Semantic social network Within the semantic web, social applications try to include semantic web technologies and best practice as FOAF, SKOS, SIOC, etc. Among applications that follow this trend, a mobile social semantic web framework, which weaves a distributed social network based on semantic technologies [11], An architecture for an open, distributed social network, which is built solely on semantic web standards [12], a social networking federation as a paradigm where information on various social network systems can be seamlessly integrated in order to provide users a uniform and semantic view of their social connections [13]. Despite all its benefits, semantic web technologies assume that users are required to have some skills as the creation of a WebID7 or a FOAF profile. In our work, we
CALBA [14] a framework designed for TSNs to select vendors as advertising sources for mobile users in a specific place. The selection is based on geographical proximity and user’s preferences with the constraint that the total advertising frequency of the selected vendors should not exceed the user specified capacity. Bfriend [15] is a location-aware ad-hoc social networking platform based on the Facebook social graph. Users receive push notification when a friend is in the same place at the same time. Snapchat [16] is a mobile messaging app for sharing pictures that disappear: it sends photos and then deletes them from the receiver’s phone a few seconds after they are viewed. The user’s content on Snapchat does not live online forever. Temporary social network is a growing field. Many other commercial temporary social media platforms have been developed as Gryphn8, Burn Note9, Wickr10, Silent circle11, LobbyFriend12, Fast society13. All offer the ability to have a private and spontaneous social experience. But existent TSN does not and cannot entirely live up to this claim, giving users a false sense of security because there are some loopholes. For example recipients can take a screenshot or plugging the phone into a computer to recover the content from the local memory. C. Location based social network / services There are various definitions such as the one proposed by Zheng et al. [17] “A location-based social network (LBSN) does not only mean adding a location to an existing social network so that people in the social structure can share location-embedded information, but also consists of the new social structure made up of individuals connected by the interdependency derived from their locations in the physical world as well as their location-tagged media content, such as photos, videos, and texts“. Most of known social networks integrate location based-services like Facebook Places14, Foursquare15, Skout16. Besides commercial services researches have been done, targeting these kind of services we can name: Tutorial on location-based social networks [17], Moss: mobile social spaces [18], GeoLife : a collaborative social networking 8
http://armortext.co/
9 https://burnnote.com/ 1
http://www.foaf-project.org/
10 https://www.mywickr.com/en/index.php
2 http://sioc-project.org/
11 https://silentcircle.com/
3
12
4
13 http://fastsociety.com/
http://www.w3.org/2004/02/skos/ http://www.w3.org/TR/owl-time/ 5 http://georss.org/Main_Page 6 http://www.w3.org/Submission/2004/SUBM-OWL-S-20041122/ 7 http://www.w3.org/wiki/WebID
14
http://www.lobbyfriend.com/
www.facebook.com/about/location https://fr.foursquare.com/ 16 http://www.skout.com/ 15
service among user, location and trajectory. [19]. Most of proposed solutions take account users preferences, policies and ignore certain dimension like communities and other user context aspect like the execution context.
The Communities Manager is the coordinator, which interacts with the ontologies. The core of this manager is a set of services used to supply and to extract knowledge from ontologies. This manager may be centralized or distributed.
D. Community-awareness application The concept of communities is common to different domains such as social sciences, social networks, knowledge management, learning, etc. The number of communities has skyrocketed with the growth of social networks. Researches are more interested in communities discovering as [20]. Their challenge is to cluster users based on their affinity to retrieve implicit communities. In pervasive social computing, applications are increasingly oriented communities and provide service for this granularity of users. Beside mainstream service (e.g. Google Group17, Yahoo Groups18), others have been proposed by academic research as Group knowledge management for context-aware group applications and services [21]. Pervasive CSCW for smart spaces communities [22]. Towards a Context Aware Mobile Community Application Platform [23]. Although, the existing applications offer services and information related to a specific theme, which has to be defined in advance by developers. Therefore, community application represents new challenge in terms of the creation of spontaneous community and the use of new features that make easy for the user to find documents, people, or places.
The Member Device allows the user to discover and interact with the communities around. Two kinds of services can be deployed on the user device:
We presented the four dimensions that will be merged in our application while adding a layer of context awareness. Taldea tries to remove some of cited limitations and provide a range of pervasive content and services that foster creation of spontaneous communities and exchange of information. III.
TALDEA ARCHITECTURE
Taldea is based on a hybrid architecture; it shares properties with both centralized and P2P architectures and takes advantage of both approaches.
Fig. 1. An Overview of Taldea’s Architecture.
The architecture contains two types of structural entities: Communities Manager and Member Device. Figure 1 shows an overview of the Taldea Architecture:
-
-
A. The Communities Manager The core of this manager is a set of services for supplying and extracting knowledge from ontologies and updating them. It interacts with two types of ontologies: the communities’ ontology and the domain ontology. The communities’ ontology is used to formally describe the communities’ resources. We consider as resources the description of the community itself and all entities involved in the life cycle of a community such as user, multimedia document and service. The domain ontology describes a specific domain with concepts and relationships that exist between those concepts; in our case is the botanical domain. The Communities Manager handles the knowledge base. Among its functionalities are listed: - Generate requests; - Query ontologies; - Annotate resources; - Populate the ontologies; - Compute the semantic distance between concepts; - Hold Policies for services deployment; - Hold user rules. This manager may be centralized or distributed among a set of user devices depending on the availability of resources. 1) Annotation Processor: The purpose of the annotation process is to give meaning to the user queries. With Taldea, the user can enter queries in natural language to search for resources. The query and the resource description will be annotated through TextAnnot19. Based on the result of the annotation, we perform a SPARQL query. 2) Query Engine: Usually, a non-expert user is not able to query ontology. We need to transform any natural language query to SPARQL20 query to search for communities’ resources. The user query will be annotated through the Annotation Processor. Using the results of the annotation, a SPARQL query is formulated with the concepts results in the 19
17 https://plus.google.com/getstarted/suggest?fww=1 18 https://fr.groups.yahoo.com/neo
Core services: basic services including user login to the application, access to communities and their common space. These services interact with the Communities Manager in an explicit way by user request or user publishes, or in implicit based on user-context. Community-aware services: they facilitate information exchange and communication. They are deployed according to the social context of the user that corresponds to community context.
A web service developed by our research team, it provides information in the form of some annotations based on botanic ontology. (http://themat2i.univ-pau.fr:8080/TextAnnot-WWW/annotation.jsp) 20 http://www.w3.org/TR/rdf-sparql-query/
clause Where. Therefore, Query Engine module allows users to query the community knowledge base without using a complex query language. TABLE I.
EXAMPLE OF ANNOTATED QUERY
I'm looking for orchid community TABLE II.
B. The Member Device Taldea is an application running on mobile devices. It provides a set of services allowing the user to join and exchange information within communities. We have classified services in two groups according to how they will be deployed: core services, communities related services. 1) Core services: Core services are mandatory for user connection and access to the communities. Regardless of user or community type, these services must be automatically (at the runtime) deployed. Figure 2 describes the interactions between these user’s side services and the communities’ manager.
EXAMPLE OF QUERY FORMULATION
PREFIX onto : Select ?community Where { ?community rdf:type onto:Communities. ?community onto: hasInterest onto:"orchid ". }
3) Similarity Measurement: In the field of information retrieval, semantic similarity measures are used to assess the semantic proximity between a query and a document. In our context of community applications, we evaluate the semantic similarity between the user’s profile and communities’ description, by the mapping between the user’s interests and community’s interests in the domain ontology (i.e. the botanic ontology). We have adopted an approach based on arcs between concepts because other approaches used the frequency of information. This is not significant in the case of community because it is created around infrequent concepts in the corpus. We choose the measure of Wu and Palmer [24]. It is simple to implement, has good performance compared to other similarity measures. The Measure of Wu and Palmer is a measure between concepts in ontology. Wu and Palmer, similarity metric measures the depth of the two concepts in the ontology, and the depth of the least common subsumer (LCS), and combines these figures into a similarity score (1). The key idea of communities’ recommendation is to compute semantic similarity between the user’s Interest (UI) and the community’s Interest (CI). (1)
4) Policies rules Engine: Popular classification of groups of users is not based only on the interest but also on the social relationships with other members like (Friend, Family and Coworkers, etc.). In this case, the user can define policies for a specific group. Most existing applications enable users to restrict and authorize the shared content but rarely the applications permit to control the shared services between community members. When the relationship with other members is more significant than the common interest, reasoning with the user-defined policies will manage unforeseen situations. Taldea cracks privacy issues by proposing a policies rules engine that deals with user policies to manage the shared content and the shared services.
Fig. 2. Access to communities.
a) Community Access: This service facilitates the user access to communities. There are three ways to access a community in Taldea: recommendation, search and creation of a community. As shown in the figure 2 above, (1) The user logs in, (2) her interests are mapped to the botanical ontology, as well as the interests of existing communities (3). Then, similarity measures are used in order to identify communities semantically similar to the user's profile. If the user is not satisfied with the proposals, (4) Taldea, allows her to enter a query in natural language for searching for other communities. (5) The query is annotated through the Annotation processor. (6) A SPARQL query is formulated from annotation results to query the ontology (12). (7) If the user is not satisfied with the search results, (8) she can create a spontaneous community. (9) Inference rules that can be used to infer others information as the type of community. Finally, the community ontology is instantiated with the user-entered information and the inferred information (12). b) Community Space: A core part of Taldea is the community space. The prerequisites for Taldea space are selfmanagement, update, and privacy policies compliance. It includes several functionalities such as viewing, publishing and searching content. Taldea ensures knowledge base management and enables user to consult, publish, search, and annotate content by accessing to the common space of the community. This service is independent from the community type but its content changes from one community to another. 2) Community related Services: Due to the limited resources of mobile devices and the growing number of services, context awareness becomes a major requirement in
pervasive environment. Applications must be adapted to context constraints (such as the host’s resources, the services properties and the surrounding environments) and must provide appropriate services. The discovery of services is a promising approach to address the challenge. Many approaches have been proposed in the field of semantic web service discovery; they are based on descriptions of web services (OWL-S21, WSDL-S22, WSMO23). Most approaches consider other dimensions of context such as the environment, user, location, temporality, etc. In our case, we consider the social dimension. The problem of matching service advertisement and community profile is transformed on the computation of semantic similarity between concepts in the botanical ontology. For this purpose, we use the OWL-S ontology in order to describe the community related services. To allow the semantic search of services, we use the classical work of [25]. The main idea is to match the community description with the service description. The service‘s description is enhanced by annotating with interests of communities that have used this service. We recall that community interests are concepts from the domain ontology. For a service concept Sc and community concept CC, we consider the match as exact if Sc is equivalent to CC; plugin-in if Sc subsumes CC; subsumes if Sc is subsumed by CC, fail otherwise (cf. table (a)). For example, a created community has as interest orchid. The available services are annotated with concepts of communities that have used. Annotations are illustrated in table (b). In table (c), we present the result of matching the interest of the community and the annotations of available services. TABLE III.
(a)
Each device deploy the selected services via Kalimucho, a service-based reconfiguration platform that implements the deployment strategy [10]. IV. IMPLEMENTATION & POTENTIAL USE The proposed application intends to help users in the creation of an ephemeral community in a specific place. Several scenarios highlight the importance of ephemeral community. Several use cases can be defined for Taldea like Conference, Exposition, Festival, Sport Event, Natural Disaster, etc. Let us imagine the use case related to our application context, the Mosaïc which is a botanical park, located near the French city of Lille. In this park, we can find a variety of stakeholders (visitors, gardeners, eco-guards, external participants, etc.) with mobile devices. Screenshots in figure 3 shows some implemented features. A visitor equipped with her Smartphone, enters into the park. The Smartphone automatically connects to Taldea using Geofencing24 features. The user has indicated on her profile that she is fan of Perennial plant (a). Once the user is connected, she can join one or more communities. The community access module includes three functions: recommendation, research and creation of communities (b). To access a community, it is recommended for a new user to follow these steps.
MATCHING EXAMPLE BETWEEN SERVICES DESCRIPTION AND COMMUNITIES DESCRIPTION. Logical description
Sim (CC, Sc)
(b) CommunityInterest Service 1 Service 2 Service 3
Meaning
Value
CC ≡ Sc
Exact
1
Sc ⊆ CC
Plugin-in
2/3
CC ⊆ Sc
Subsume
1/3
CC ≠ Sc
Fail
0
Concepts Orchid Flower Tree, Plant Ophrys apifera
(c) Service 1 Service 2 Service 3
(a) (b) Communities that are semantically related to Perennial plant will be automatically recommended to the visitor (c). Moreover, she can search for a specific community by formulating a natural language query (d). For instance, let us suppose that the visitor connects to the perennial community of interest ‘fans of Perennial plant ’.
Community Interest 2/3 0 + 2/3 1/3
Community related services are services that will be useful to the user to exchange information and communicate. A nonexhaustive list of services provided by Taldea is (Geofincing, PointsOfInterests, Compass, Tracks, P2P Communications, MediaPlayer, MediaRecorder, PictureTake, PictureDisplay, etc.). 21
http://www.w3.org/Submission/2004/SUBM-OWL-S-20041122/ 22 http://www.w3.org/Submission/WSDL-S/ 23 http://www.w3.org/Submission/WSMO/
(c) 24
(d)
The geo-fencing approach is based on the observation that users move in a virtual perimeter for a real-world geographic area. It allow us to track people on a mobile subscriber list based on proximity to a particular retail store and sending them tailored messages relating to that store.
During her tour in the botanical garden, she identifies some rare 'Ophrys apifera' flowers. She first takes pictures of her discovery. She creates an ephemeral community named “Help rare orchids” (e). The shared picture in the community space is annotated with spatiotemporal data in order to facilitate the location of flowers (f).
TABLE IV.
SEMANTIC DISTANCE USING WORDNET AND BOTANICAL ONTOLOGY
SimWU&PALMER (Tree,Flower) SimWU&PALMER (Gladialus,Pink)
Wordnet 0.7826 0.1905
Botanical ontology 0.6667 0.2857
On the one side, Tree and Flower are concepts semantically close in general context but in botanical context they are farther. On the other side, the distance value between Gladialus and pink is higher in the botanical ontology because the pink color is frequently associated to these flowers. Unsurprisingly, for a given domain, semantics is richer into a specialized ontology. For performance evaluation, we measure two indexes: Precision and Recall. Precision is the fraction of the found items that the requesting users consider relevant; Recall is the fraction of the relevant items that has been found. Precision and recall are formally defined as follows:
(e) (f) The created community is recommended to relevant users of Taldea (they can be members of other communities) currently present and visiting the park. A short time later, the creator received several answers. Interested people moved on the scene to discover or to admire these flowers. Once connected to a community, the user can view the community space (f). She can access to the list of related services (g) as P2P chat (h).
(g)
(h)
(2)
Recall
(3)
This experiment results show the curves recall and precision by changing the ontology on which we calculate the similarity. We have evaluated our approaches using the semantic similarity of: Wordnet25 and TextAnnot26. We suppose that a user X has in his profile the perennial flowers. We will recommend communities and potentially services related to his interest. In the first step of experiments we used Wordnet to evaluate the effectiveness of the community and services results. In the second stage, we evaluate the two approaches with TextAnnot. To evaluate the results, we parse items noting 1 if the item is pertinent for the user, 0 otherwise. For each result, we calculate the precision and the recall of the first 5 returned items. The evaluation of communities’ recommendation in Figure 4 shows that the use of a specific ontology has best result than the use of generic ontology. On the other side, Figure 5 shows opposite results for services discovering.
Fig. 3. Screenshots of the Taldea application running on a Samsung phone showing some implemented features such as the communities access, the chat, the community Space.
Taldea consist of an android application. It is composed of Kalimucho Business Components (BC) [10]. So before installing Taldea, the Kalimucho platform is required on the user devices. The deployment of the application is performed at run time and can be modified while the application is running without stopping it.
Fig. 4. Communities recommandation
V. PRIMARY EXPERIMENTATION In annotation field, two kinds of ontology are generally used: general ontologies shared by all the domains, and specialized ontologies which represent exhaustive information about a domain in accordance with the document corpus or a specific uses. TextAnnot is available online and connected to a botanical ontology and corpus. It will be soon configurable to take account any ontology and corpus. Table 4 shows a comparison between the results of computing semantic distance between concepts in a general ontology and in a specialized ontology.
Fig. 5. Services discovering
25 26
http://marimba.d.umn.edu/ http://themat2i.univ-pau.fr:8080/TextAnnot-WWW/annotation.jsp
Although, we note that the difference between the two curves for each case is not very high. The experimentation was made for a single query but if we calculate the average on a set of queries, the curves tend to be smother and to get closer.
[9]
[10]
VI. CONCLUSION In this paper, we have presented Taldea a communitycentered application that helps users to access to communities and organize social exchanges between users in a geographic territory. The key to design such social mobile application is considering communities as distinct social entities that should be supported with services as single user. For this purpose, we assume a service oriented application paradigm and we apply policies for the services deployment. Likewise we aimed at showcasing the benefits of using the semantic web standards, technologies and best practice in social mobile. Our immediate plan is to evaluate the Taldea usability by assessing the user’s satisfaction for social exchange against other community applications. In future work, several issues will be investigated. We plan to integrate spatial and temporal dimensions for community recommendation and enrich the community description to include a spatio-temporal contextualization of social exchanges between users.
[11]
[12]
[13]
[14]
[15]
[16]
[17]
REFERENCES [1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
C.-m. A. Yeung, I. Liccardi, K. Lu, O. Seneviratne, and T. Berners-Lee, "Decentralization: The future of online social networking," in W3C Workshop on the Future of Social Networking Position Papers, 2009. C. J. Carpenter, "Narcissism on Facebook: Self-promotional and antisocial behavior," Personality and individual differences, vol. 52, pp. 482-486, 2012. K. Graffi, C. Gross, D. Stingl, D. Hartung, A. Kovacevic, and R. Steinmetz, "LifeSocial. KOM: A secure and P2P-based solution for online social networks," in Consumer Communications and Networking Conference (CCNC), 2011 IEEE, 2011, pp. 554-558. S. Buchegger, D. Schiöberg, L.-H. Vu, and A. Datta, "PeerSoN: P2P social networking: early experiences and insights," in Proceedings of the Second ACM EuroSys Workshop on Social Network Systems, 2009, pp. 46-52. N. Al Mutawa, I. Baggili, and A. Marrington, "Forensic analysis of social networking applications on mobile devices," Digital Investigation, vol. 9, Supplement, pp. S24-S33, 2012. M. S. Lam, "How mobile disrupts social as we know it," presented at the Proceedings of the 2013 international conference on Intelligent user interfaces, Santa Monica, California, USA, 2013. D. Schuster, A. Rosi, M. Mamei, T. Springer, M. Endler, and F. Zambonelli, "Pervasive social context: Taxonomy and survey," ACM Transactions on Intelligent Systems and Technology (TIST), vol. 4, p. 46, 2013. O. Coutand, O. Droegehorn, K. David, P. Nurmi, P. Floréen, R. Kernchen, S. Holtmanns, S. Campadello, T. Kanter, and M. Martin, "Context-aware group management in mobile environments," in IST Mobile Summit, 2005.
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
G. Ben Nejma, P. Roose, J. Gensel, M. Dalmau, and M. Ghorbali, "Taldea: A Tool for Fostering Spontaneous Communities," in Current Trends in Web Engineering. vol. 8295, Q. Sheng and J. Kjeldskov, Eds., ed: Springer International Publishing, 2013, pp. 293-301. Keling DA, Marc DALMAU, Philippe ROOSE, "Kalimucho: Middleware for mobile applications," in ACM SAC 2014 –ACM Press – 24-28/03, Gyeongju, Korea, 2014, pp. 413-419. S. Tramp, P. Frischmuth, N. Arndt, T. Ermilov, and S. Auer, "Weaving a distributed, semantic social network for mobile users," in The Semantic Web: Research and Applications, ed: Springer, 2011, pp. 200-214. S. Tramp, P. Frischmuth, T. Ermilov, S. Shekarpour, and S. Auer, "An architecture of a distributed semantic social network," Semantic Web, 2011. W. Chao, Y. Guo, and B. Zhou, "Social networking federation: A position paper," Computers & Electrical Engineering, vol. 38, pp. 306329, 2012. W. Xu, C.-Y. Chow, and J.-D. Zhang, "CALBA: capacity-aware location-based advertising in temporary social networks," in Proceedings of the 21st ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, 2013, pp. 354-363. V. Smailovic and V. Podobnik, "Bfriend: Context-aware ad-hoc social networking for mobile users," in MIPRO, 2012 Proceedings of the 35th International Convention, 2012, pp. 612-617. N. A. Poltash, "SNAPCHAT AND SEXTING: A SNAPSHOT OF BEARING YOUR BARE ESSENTIALS," Rich. JL & Tech., vol. 19, pp. 14-14, 2013. Y. Zheng, "Tutorial on location-based social networks," in Proceedings of the 21st international conference on World wide web, WWW, 2012. A. V. Zhdanova, M. Davies, O. Jorns, V. Stavroulaki, P. Demestichas, M. Gonzalez, K. Moessner, F. Carrez, H. Rajasekaran, and L. L. Iacono, "MOSS: Mobile Social Spaces," in 1st International Workshop on Blending Physical and Digital Spaces on the Internet (OneSpace2008) at the 1st Future Internet Symposium (FIS’08), Vienna, Austria, 2008. Y. Zheng, X. Xie, and W.-Y. Ma, "GeoLife: A Collaborative Social Networking Service among User, Location and Trajectory," IEEE Data Eng. Bull., vol. 33, pp. 32-39, 2010. D. Leprovost, L. Abrouk, and D. Gross-Amblard, "Discovering implicit communities in web forums through ontologies," Web Intelligence and Agent Systems, vol. 10, pp. 93-103, 2012. H. De Silva, K. Moessener, and F. Carrez, "Group knowledge management for context-aware group applications and services," in Personal, indoor and mobile radio communications, 2009 IEEE 20th international symposium on, 2009, pp. 2851-2855. C. Lima, D. Gomes, and R. Aguiar, "Pervasive CSCW for smart spaces communities," in Pervasive Computing and Communications Workshops (PERCOM Workshops), 2012 IEEE International Conference on, 2012, pp. 118-123. C. Menkens, "Towards a context aware mobile community application platform," in Information Technology: New Generations, 2009. ITNG'09. Sixth International Conference on, 2009, pp. 1504-1509. Z. Wu and M. Palmer, "Verbs semantics and lexical selection," in Proceedings of the 32nd annual meeting on Association for Computational Linguistics, 1994, pp. 133-138. M. Paolucci, T. Kawamura, T. R. Payne, and K. Sycara, "Semantic matching of web services capabilities," in The Semantic Web—ISWC 2002, ed: Springer, 2002, pp. 333-347.
