Jungwon Cho

Personalization method for tourist point of interest (POI) recommendation

Today, travelers are provided large amount information which includes Web sites and tourist magazines about introduction of tourist spot. However, it is not easy for users to process the information in a short time. Therefore travelers prefer to receive pertinent information easier and have that information presented in a clear and concise manner. This paper proposes a personalization method for tourist Point of Interest (POI) Recommendation.

An Intelligent Course Recommendation System

Previous academic administration management systems had migrated from wired to wireless technology but were restricted to specific equipment, as such systems were not based on industry standards. The course recommendation system plays a significant role in managing the curriculum and counseling students on academic matters, with a view to fostering their academic progress. However, the system does not have the time to advise individual students on the details of which fields and courses they should pursue. The rapid progress of IT technologies has enabled users to access ?any service, anytime, anywhere?, and wireless internet services have enabled users to access internet services even while traveling. Cellular phones have evolved to smart phones, which can provide visual telephone service, DMB service, and act as smartphones. Users can access service categories that offer new possibilities and are accessed and used in ways different from traditional services. Thus, cellular phones are a representative terminal for ubiquitous learning. This paper proposes a mobile course recommendation system to help students choose and access courses necessary for their major fields of study. When students apply to take courses, the proposed system recommends the most suitable ones, using an inference engine that considers not only course sequences but also students? information. The system is able to assist the course coordinator in counseling students. The students use their personal cellular phones to track their courses and receive course recommendations from the system. We searched for relationships between subjects, using association rules, by mining data about the courses already taken by students, and compared these to existing course trees. This information could be used for updating course trees.

News video retrieval using automatic indexing of korean closed-caption

Knowledge-based video retrieval is able to provide the retrieval result that corresponds with conceptual demand of user because of performing automatic indexing with audio-visual data, closed-caption, and so on. In this paper, we present the automatic indexing method of Korean closed-caption for knowledge-based video retrieval and the retrieval scheme using the indexed database. In the experiment, we have applied the proposed method to news video with the closed-caption generated by Korean stenographic system, and have empirically confirmed that the proposed method could provide the retrieval result that corresponds with more meaningful conceptual demand of user.

Sensor fusion for accurate ego-motion estimation in a moving platform

With the coming of “Internet of things” (IoT) technology, many studies have sought to apply IoT to mobile platforms, such as smartphones, robots, and moving vehicles. An estimation of ego-motion in a moving platform is an essential and important method to build a map and to understand the surrounding environment. In this paper, we describe an ego-motion estimation method using a vision sensor that is widely used in IoT systems. Then, we propose a new fusion method to improve the accuracy of motion estimation with other sensors in cases where there are limits in using only a vision sensor. Generally, because the dimension numbers of data that can be measured for each sensor are different, by simply adding values or taking averages, there is still a problem in that the answer will be biased to one of the data sources. These problems are the same when using the weighting sum using the covariance of the sensors. To solve this problem, in this paper, using relatively accurate sensor data (unfortunately, low dimension), the proposed method was used to estimate by creating artificial data to improve the accuracy (even of unmeasured dimensions).

Map Representation for Robots

Map-building and localization for robots are the most basic technology required to create autonomous mobile robots. Unfortunately, they are difficult problems to comprehensively handle. If expensive sensors or a variety of external devices are used, then the problems can be resolved. However, there are still limits for various environments or platforms. Therefore, many researchers have proposed various different methods over a long period of time, and continue to do so today. In this paper, we first look at the state of existing research for map representations used in map-building and localization. We divide them into four main categories and compare the differences between them. These identified properties between the four categories can be used as good standards for choosing appropriate sensors or mathematical models when creating map-building and localization applications for robots.

Personalized Curriculum Recommender System Based on Hybrid Filtering

Recently, the teaching-learning paradigm is focusing on learners. The individual’s right to select curriculum is gaining ground. As this selection right increases, there is increasing concern and more time needs to be invested in selecting the curriculum suitable for an individual’s situation and preferences. Therefore, an individualized service that can recommend a desirable curriculum to individuals is needed to minimize individuals’ efforts and help them make the right choices. This paper proposes a curriculum recommender system through which individual learners can get advice when they enroll. This research provides the foundation of learner-oriented education by providing a personalized curriculum from the beginning of a course of study.

A Personalized Assessment System Based on Item Response Theory

Computerized adaptive testing (CAT) is a method of administering tests that adapts to the examinee’s ability level. Previous research has focused on estimating the examinee’s ability accurately and on providing adequate feedback upon analyzing the examinee’s ability. However, in order for students to use the feedback, they must find courses or learning materials themselves. It is difficult to make customized learning available continuously. Therefore, we used adaptive testing to estimate a student’s ability and to identify a number of student characteristics. This paper recommends content that can reinforce areas in which the student needs improvement. We applied our system at an actual education site. The group that used our recommendation module learned more effectively than the control group. By using this system, teachers will be able to monitor students closely. This enables customized learning which allows students to study effectively without necessitating the effort to search for learning materials. Customized learning will increase interest in.

A framework for online gait recognition based on multilinear tensor analysis

The gait recognition is to recognize an individual based on the characteristics extracted from the gait image sequence. There are many researches for the gait recognition which use diverse kinds of information such as shape of gait silhouette, motion variation caused by walking, and so on. In general, shape information is more useful for recognition. However, shape information is influenced by a variety of factors, which degrade the recognition performance. Moreover, the information used in most of those studies might be able to be extracted after all of one or more sequences of the gait cycle are known. And it is also hard to discriminate the gait cycle from given gait sequences exactly by the online approach. In regard to these difficulties, we propose a novel gait recognition method based on the multilinear tensor analysis. To recognize the cyclic characteristic of gait without an exact division for the gait cycle, this paper’s propose is the method to form the accumulated silhouette and then describes those as the tensor. For the accumulated silhouette proposed by this paper, the image sequence of one gait cycle is divided into four sections in the training phase. However, discrimination for the gait cycle in the training phase is not directly related to the recognition phase, thus the online approach is possible. We first form the accumulated silhouettes for every individual using gait silhouettes within each section. And then, we represent these accumulated silhouettes as the tensor. Using a multilinear tensor analysis, we compute the core tensor which governs the interaction between factors organizing the original tensor, and then compose the basis to recognize the individual in the online recognition framework. Finally, we recognize the individual using the computation of similarity based on the Euclidean distance, which is more suitable to our method. We verify the superiority of the proposed approach via experiments with real gait sequences.

Personalized item generation method for adaptive testing systems

An Intelligent Tutoring System (ITS) must provide suitable feedback to learners based on tests adapted to the learners’ ability levels. An ITS selects the item and content based on what it knows about the learners from previous items. Previous research has focused on estimating a learner’s ability accurately or providing adequate feedback based on an analysis of the learner’s ability. However, it is often difficult to make customized learning continuously available in ITSs. In the present study, we used adaptive testing to estimate a learner’s ability and to determine a number of learner characteristics to create a learner profile. This method selects items and creates a customized assessment sheet for adaptive testing that considers both the learner’s level and characteristics. The proposed method assesses a learner’s weak subject areas and item types by studying available information and analyzing individual abilities to guide learners as to which fields of study would be suitable and which courses they should take. We tested our customized learning module at an actual educational institution. The group that used our recommendation module learned more effectively than the control group (the mean test scores of the group that used the module were high and the deviations were low). Using the learner model, teachers will be able to analyze learners in detail, enabling customized learning that allows learners to study effectively without requiring a great effort to search for learning materials. Customized learning will increase interest in learning and understanding.

Collaborative Project Curriculum applying Project-based Learning

In diverse professional fields including information and communication field, there is huge gap between required practical ability in industrial and that of graduated students actually learn in university. In this paper, we propose collaborative project curriculum applied to project-based learning in order to train affordable specialists who have a systematic problem-solving ability in industrial site. Applying this curriculum to The Cultural Technology Education Center of Jeju National University, we confirm that our curriculum improves ability of students to perform team project systematically such as skills in paper writing, presentation speech, problem solving, and collaborative work. Students are also reported high satisfaction about the proposed curriculum. Therefore, we expect that our curriculum will successfully be applied to other fields.