Nammee Moon

Recommender system design using movie genre similarity and preferred genres in SmartPhone

As e-commerce (e.g. www.amazon.com) and social media (e.g. www.facebook.com) services evolve, studies of recommender systems advance, especially concerning the application of collective intelligence to personalized service. With the development of smartphones and the new mobile environment, studies of customized services increase despite the physical limitations of mobile devices. A typical example combines customized services with location-based services. In this study, we propose a recommender system using movie genre similarity and preferred genres. A movie genre similarity profile is designed and generated to provide related services in a mobile experimental environment before prototyping and testing with data from MovieLens. In order to accomplish this, genre similarity correlations are determined with a Pearson correlation coefficient, and similar clusters are derived. The correlations within clusters are used to define genre similarity. Genre similarity is then used to recommend new genres to targeted customers.

User-selectable interactive recommendation system in mobile environment

Mobile devices need to provide more accurate and personalized information in a computing environment with a small screen and limited information retrieval functions. This paper presents a user-selectable recommendation system that reflects a user interest group by employing collaborative filtering as technique to provide useful information in a mobile environment. We form similar groups by simultaneously considering a user’s information preferences and demographics. Then we reconstruct lists of a final recommendation based on what search results the similar demographic group has chosen. This is an optional filter for the search results. This means that we provide an interactive flexible recommendation list that considers a user’s intent more actively, rather than unilaterally. We show the Mean Absolute Error result to evaluate the recommendation and finally show the realization of a prototype that is based on both the iPhone and Android phone environments.

A Cultural Dimensions Model based on Smart Phone Applications

Abstract —One of the major factors influencing the phenomenal growth of the smart phone market is the active development applications based on open environments. Despite difficulties in finding and downloading applications due to the small screens and inconvenient interfaces of smart phones, users download applications nearly every day. Such user behavior patterns indicate the significance of smart phone applications. So far, studies on applications have focused mainly on technical approaches, including recommendation systems. Meanwhile, the issue of culture, as an aspect of user characteristics regarding smart phone use, remains largely unexamined throughout the world. Hence, the present study attempts to analyze the highest ranked smart phone applications downloaded and paid for that are ranked the highest in 10 countries (Korea, Japan, China, India, the UK, USA, Indonesia, Canada, France, and Mexico) and we then derive the CDSC (Cultural Dimensions Score of Content) for these applications. The results derived are, then, mapped to the cult ural dimensions model to determine the CISC (Cultural Index Score for Country). Further, culturally significant differences in smart phone environments are identified using MDS analysis.

Recommender System Using the Movie Genre Similarity in Mobile Service

Users of the various opinions and knowledge are generated and shared through the collective intelligence, a research on recommender systems are being continued at a variety of areas to use this at a personalized service. Also, despite the constraints of mobile device, personalized service is accelerating as development of the mobile environment. Therefore, we propose the recommender system using the genre similarity and preferred genre. After finding the relationship between genres by Pearson correlation coefficient, it produces a group by K-Means clustering. It creates a genre similarity profile by similar relationship between genres within a group. Suggest recommender system is reflected the genre similarity and preferred genre by target customer preferred genre. After designing and prototyping this to be able to be serviced at mobile experiment environment, it evaluates by applying to MovieLens Data set.

Towards a cultural user interface generation principles

As ubiquitous computing, and pervasive computing technology are being applied rapidly to the service industry, in the field of HCI, more complex and in-depth research are required at the moment. The efforts to make user experience more valuable using useful multimedia technologies around us, are being made in various parts of the world. In addition to this, generation studies for developing easier and faster UI are being conducted in the field of UI. Until now, the studies have been focused on generation studies using UI pattern models, but smart UI generation development, considering the cultural environment of users, is required. The purpose of this study is to suggest 3-D integrated design principles of modeling, design, and system for the design of cultural user interface generation reflecting the users’ potential culture models. For this, a CTT model was designed based on MB-UID; and a cultural UI generation system architecture was designed by using cultural dimension models and cultural markers. To analyze the cultural factors, CISC (Cultural Index Score for Country) was derived in the process of culture profiling. Then, to verify the design model, culture UI was implemented, targeting the category UI of smartphone app stores.

Object Association and Identification in Heterogeneous Sensors Environment

An approach for dynamic object association and identification is proposed for heterogeneous sensor network consisting of visual and identification sensors. Visual sensors track objects by a 2D localization, and identification sensors (i.e., RFID system, fingerprint, or iris recognition system) are incorporated into the system for object identification. This paper illustrates the feasibility and effectiveness of information association between the position of objects estimated by visual sensors and their simultaneous registration of multiple objects. The proposed approach utilizes the object dynamics of entering and leaving the coverage of identification sensors, where the location information of identification sensors and objects is available. We investigate necessary association conditions using set operations where the sets are defined by the dynamics of the objects. The coverage of identification sensor is approximately modeled by the maximum sensing coverage for a simple association strategy. The effect of the discrepancy between the actual and the approximated coverage is addressed in terms of the association performance. We also present a coverage adjustment scheme using the object dynamics for the association stability. Finally, the proposed method is evaluated with a realistic scenario. The simulation results demonstrate the stability of the proposed method against nonideal phenomena such as false detection, false tracking, and inaccurate coverage model.

Teaching-Learning Activity Modeling Based on Data Analysis

Numerous studies are currently being carried out on personalized services based on data analysis to find and provide valuable information about information overload. Furthermore, the number of studies on data analysis of teaching-learning activities for personalized services in the field of teaching-learning is increasing, too. This paper proposes a learning style recency-frequency-durability (LS-RFD) model for quantified analysis on the level of activities of learners, to provide the elements of teaching-learning activities according to the learning style of the learner among various parameters for personalized service. This is to measure preferences as to teaching-learning activity according to recency, frequency and durability of such activities. Based on the results, user characteristics can be classified into groups for teaching-learning activity by categorizing the level of preference and activity of the learner.

Towards Cultural User Interface Generator Principles

The HCI research field needs more complex and in-depth studies. Nowadays, many researchers are focusing on the user experience process that can make usability valuable. In addition to that, user behavior patterns, social interaction, and cross cultural issues are still important parts of user interface technology. On the strength of these trends, research about the interface generator has been conducted in the user interface area to make development faster and easier. This study aims to establish design principles of the cultural user interface (CUI) generator and design architecture so that culture can be reflected in the user interface generator. A CUI generator based on design principles will be adapted to multiple devices such as the web, mobile phones, IPTV and so on. The cultural dimensions model and culture markers are adopted as a method at the same time.

Trajectory based database management for intelligent surveillance system with heterogeneous sensors

In this paper, we present a database management scheme for an intelligent surveillance system utilizing multiple visual sensors and RFID readers. The objects are tracked and identified by multiple visual sensors and RFID readers. We define three different types of data structure to consistently store data for effective data storage. They contain global object number and identification as the common information of the same object. The global object number is uniquely assigned for each track object. The previously stored data without the common information is back-annotated when it is available in the system. Moreover, when the global object number changes because of imperfect detection and tracking, the system maintains consistency information between global object numbers for the same object by comparing their local target information or positions. The fragmented information for an object is also stitched through map information. The simulation results demonstrate that the database information for objects is successfully recovered with consistency.

Acoustic sensor-based multiple object tracking with visual information association

Object tracking by an acoustic sensor based on particle filtering is extended for the tracking of multiple objects. In order to overcome the inherent limitation of the acoustic sensor for the simultaneous multiple object tracking, support from the visual sensor is considered. Cooperation from the visual sensor, however, is better to be minimized, as the visual sensors operation requires much higher computational resources than the acoustic sensor-based estimation, especially when the visual sensor is not dedicated to object tracking and deployed for other applications. The acoustic sensor mainly tracks multiple objects, and the visual sensor supports the tracking task only when the acoustic sensor has a difficulty. Several techniques based on particle filtering are used for multiple object tracking by the acoustic sensor, and the limitations of the acoustic sensor are discussed to identify the need for the visual sensor cooperation. Performance of the triggering-based cooperation by the two visual sensors is evaluated and compared with a periodic cooperation in a real environment.