Ubiquitous Computing Systems 

Abstract: The recommendation system is one of the core technologies for implementing personalization services Recommendation systems in ubiquitous computing environment should have the capability of context-awareness In this research, we developed a music recommendation system, which we shall call C2_Music, which utilizes not only the user's demographics and behavioral patterns but also the user's context For a specific user in a specific context, the C2_Music recommends the music that the similar users listened most in the similar context In evaluating the performance of C2_Music using a real world data, it outperforms the comparative system that utilizes the user's demographics and behavioral patterns only

Abstract: The accelerometers integrated in today’s phones can be used to estimate the distance travelled from the accelerations made while walking. The placement of the sensor on the body is important to take into consideration. In this paper, the accelerations recorded with a daily-used phone in the trouser pocket were processed on a mobile device to detect steps and estimate the distance travelled. The outcome of the distance estimates shows an error of 0.05 metres per one metre and can be improved through calibration. This distance was applied in the motion model of a particle filter, and fused with a map of the building. The results establish that the estimates of the algorithm are valuable when fusing with other technologies or environment information, to aid the estimation of the location.

Abstract: —Monitoring of road surface conditions is a criticalactivity in transport infrastructure management. Many researchsolutions have been proposed in order to automatically controland check the quality of road surfaces. Most of them makeuse of expensive sensors embedded in vehicles or mainly focuson detection of specific anomalies during monitoring activity.Inthis paper, we describe the design of a system for collaborativemonitoring of road surface quality. The overall architectureencompasses the integration of a custom mobile application, ageoreferenced database system and a visualization front-end.Road surface condition is summarized through a roughnessparameter computed using signal processing algorithms runningon mobile devices. The roughness values computed are sub-sequently transmitted and stored into a back-end geographicinformation system enabling processing of aggregated tracesand visualization of road conditions. The proposed approachintroduces a thoroughly integrated system suitable for monitoringapplications in a scalable, crowdsourcing collaborative setting.Keywords–Roughness; Accelerometer; Smartphone; Monitor-ing; Cloud.

Abstract: Most social network websites rely on people's proximity on the social graph for friend recommendation. In this paper, we present MatchMaker, a collaborative filtering friend recommendation system based on personality matching. The goal of MatchMaker is to leverage the social information and mutual understanding among people in existing social network connections, and produce friend recommendations based on rich contextual data from people's physical world interactions. MatchMaker allows users' network to match them with similar TV characters, and uses relationships in the TV programs as parallel comparison matrix to suggest to the users friends that have been voted to suit their personality the best. The system's ranking schema allows progressive improvement on the personality matching consensus and more diverse branching of users' social network connections. Lastly, our user study shows that the application can also induce more TV content consumption by driving users' curiosity in the ranking process.