Nowadays mobile devices, such as smartphones, are becoming increasingly popular. A new report published by analyst firm Juniper Research forecasts that the number of global smartphone shipments will reach one billion per annum in 2016, up from 302 million in 2010 . Another market area, which is going hand in hand with the spread of smartphones, is certainly the area of mobile applications, also called mobile Apps. According to latest data published by AndroLib , the estimated number of Applications downloaded in the Android Market (an online software store developed by Google) is currently about 8 billion and each day more and more new applications are available for download.
For an average user, finding the desired resource at the right time in this wide array can be very difficult and time consuming. Mobile Recommendation is a new paradigm that sensibly increases the usability of mobile system, by proactively providing personalized and focused access . Recently, the recommendation process has been performed by using situation-awareness.
Situation-awareness is a computing paradigm in which applications can sense and explore the situation of a user in order to identify her/his demand at a certain time . The fundamental vehicle to determine the situation of a user is the context, i.e., suitable circumstance information captured from a physical or logical environment. This form of autonomous perception implies reasoning, decision, adaptation, and other features of cognitive system , as well as dealing with an intrinsic uncertainty in data .
In the literature we can find  a Situation-Aware Resource Recommender (SARR) for mobile user, based on a cognitivist approach, i.e., it is a representational system based on symbolic information processing. This solution uses a Fuzzy engine that takes advantage of a user calendar. More specifically, the user calendar acts as a reference for the parametrization of such fuzzy rules for each user. Nevertheless the use of a calendar to make a reference schedule is an explicit input required to the user. On the contrary, context information should be collected in terms of implicit input, coming from changes in the environment.
To avoid using explicit inputs as context sources, in this thesis we propose, design and implement an approach based on the emergent paradigm  for detecting events.
Emergent paradigms are based on the principle of self-organization , which means that a functional structure appears and keeps spontaneously. The control needed to achieve results is distributed over all participating entities. In the literature, the mechanism used to organize these types of systems and collective behavior that emerges from them has become also know as swarm intelligence: a loosely structured collection of interacting entities . The fact that simple individual behaviors can lead to a complex emergent behavior has been known for centuries. More recently, it has been noted that this type of emergent collective behavior is a desirable property in pervasive computing .
In this thesis we present how this form of collaborative situation awareness can be implemented by focusing on an important class of events: social events (e.g., meetings, conferences, festivals, and so on). We discuss a collaborative multi-agent scheme for the detection of such events, structured into three levels of information processing. The first level is managed by a stigmergic paradigm, in which marking agents leave marks in the environment in correspondence to the position of the user. The accumulation of such marks enables the second level, a fuzzy information granulation process, in which relevant events can emerge and captured by means of event agents. Finally, in the third level, a fuzzy inference process, managed by situation agents, deduces user situations from underlying events. The scheme is implemented through Repast, an agent-based modeling and simulation platforms.
The implemented system is tested considering representative and real scenarios, considering four different types of situation.
The scheme is tested on three synthetic scenarios and on a real scenario, considering four different types of situation. For each scenario, the scheme has proved to be able to recognize the four types of situation just approximately at the instants when these situations occur.
- Juniper Research, Smartphone Evolution Strategies: Premium, Standard and Economy Markets 2011-2016, July 2011
- Android Market statistics from AndroLib, http://www.androlib.com/appstats.aspx
- Ricci F (2011) Mobile recommender system. International Journal of Information Techonology and Tourism 12(3): 205-231
- Terveen L, Hill W (2001) Beyond recommender system: helping people help each other. In: Carroll JM (ed) Human-Computer interaction in the new millennium. Addison Wesley, New York, pp 487-509.
- Vernon D, Metta G, Sandini G (2007) A survey of artificial cognitive systems: implications for the autonomous development of mental capabilities in computational agents. IEEE Transactions On Evolutionary Computation 11(2): 151-180.
- Ciaramella A, Cimino MGCA, Lazzerini B, Marcelloni F (2010) A situation-aware resource recommender based on fuzzy and semantic web rules. International Journal of Uncertainty, Fuzziness and Knowledge-Based System 18(4): 411-430
- Ciaramella A, Cimino MGCA, Lazzerini B, Marcelloni F (2010) Using context history to personalize a resource recommender via a genetic algorithm. In: Proceedings of IEEE International Conference on Intellingent System Design and Applications (ISDA’10). Cairo, pp 965-970
- F. Heylighen, C. Gershenson, The Meaning of Self-organization in Computing, IEEE Intelligent System, Section Trends & Controversies – Self-organization and Information System, July/August 2003, pp. 72-75.
- P. Barron, “Using Stigmergy to Build Pervasive Computing Environments”, PhD Thesis in Computer Science, University of Dublin, Trinity College, October 2005