Abstract:In this paper, a recommendation system based on a mobile and web application is proposed for indoor decoration. The main contribution of this work is to apply two-stage filtering using linear matching and collaborative filtering to make recommendations. In the mobile application part, the image of the medium captured by a mobile phone is analyzed using color quantization methods, and these color analysis results along with other user-defined parameters such as height, width, and type of the product are sent to the web server. In the web application part, a large data set is first filtered via linear matching in which the color content of the medium and user-defined parameters received from the mobile application are matched to those for the products stored in the database. We then apply second-stage filtering, namely collaborative filtering, on the reduced data set. Performance evaluations of various color quantization methods and collaborative filtering methods used in the system are made. Results show the feasibility of using scalar quantization as a color quantization method and the K-nearest neighbor in the collaborative filtering method for our recommendation system. Overall evaluation of the system shows that our recommendation system provides around 90% accuracy.

Movement of Enhancing Opportunities and Improving Technology", abbreviated as Fatih, is among the most important educational projects in Turkey, in which students and teachers can use their tablet PCs to obtain educational data (text, images, media, etc.) stored in cloud servers. However, the limited network bandwidth and increase in both the number of users and their educational data sizes degrade the system performance and even negatively affect the overall usability of the system. Proxy server is a solution approach to both decreasing network traffic and increasing the efficiency in data transfers between end users (tablets) and cloud servers. In case of using a proxy server, synchronization problems arise. In this paper, distributed file synchronization approaches such as SyncML, Rsync and CouchDB have been reviewed and compared for their feasibility of employing in this school level proxy server based distributed systems architecture.

Movement of Enhancing Opportunities and Improving Technology", abbreviated as Fatih, is among the most important educational projects in Turkey, in which students and teachers can use their tablet PCs to obtain educational data (text, images, media, etc.) stored in cloud servers. However, given the large number of users along with vast amount of educational data to be accessed, traffic overload and ineffective signaling protocols to manage connection and synchronization are the potential problems in Fatih project. To provide possible solution to this, we propose to employ proxy server at school level. A signaling protocol that provides connection mechanism and synchronization between tablet-proxy server and tablet-cloud server, and that meets cloud-based SaaS (Software as a service) in this architecture is necessary. In this paper, signaling protocols such as XMPP (Extensible Messaging and Presence Protocol), SIP (Session Initiation Protocol), and ARCSPXP (ARDIC Cloud Service Platform Extension Protocol) are reviewed and compared for their feasibility of employing in this school level proxy server based architecture.

Movement of Enhancing Opportunities andImproving Technology", abbreviated as Fatih, is among the most important educational projects in Turkey, in which students and teachers can use their tablet PCs to obtain educational data (text, images, media, etc.) stored in cloud services. However, the limited network bandwidth and increase in both the number of users and their educational data sizes degrade the system performance and even negatively affect the overall usability of the system. Using proxy server is an approach to both decreasing network traffic and increasing the efficiency in data transfers between end users and cloud services. In this approach, it is critical that when synchronization process will be performed. In this paper, bandwidth utilization, CPU and memory usage information have been examined for obtaining accurate synchronization time for this school level proxy server based distributed systems architecture. In order to minimize effect of too large and too small CPU and memory usage data, we use min-max normalization. Also, multiple linear regression is used to estimate a node's availability for communication. Experimental results show that proposed technique is feasible not only for school level proxy server based systems but also for other systems.

In this study, a framework of the system which consists of mobile and web application to help user selection of decoration products like furniture, paint and carpet has been introduced. System is based on color analysis of the image of environment to be decorated and products to be selected at mobile and web applications. It also provides user to input some additional criteria. Using the color content and user requirements, the web application makes recommendations about the decoration products. This paper provides some preliminary results using scalar quantization based color content analysis and linear matching based recommendation.