Minsung Hong

Social recommendation service for cultural heritage

Cultural heritage is a domain in which new technologies and services have a special impact on people approach to its spaces. Technologies are changing the role of such spaces, allowing a more in-depth knowledge diffusion and social interactions. Static places become dynamic cultural environments in which people can discover and share new knowledge. Nowadays, cultural heritage is approaching to a new digital era in which people become active elements, as recipients of the actions ensuring the sustainability of such heritage, both moneywise but also simply as the perceived quality of life. In this perspective, this paper presents a novel recommender system to individual and people group in order to create a social recommendation service for cultural ICT applications. As key aspect of the presented work, we introduce a method for discovering and exploiting social affinity between users based on artwork features and user experience. In addition, we propose an architecture of the recommender system related with the affinity and discuss the architecture in terms of sparsity, group recommendation, and sustainability.

MyMovieHistory: Social Recommender System by Discovering Social Affinities Among Users

ABSTRACT Social network information has recently been used for the improvement of the performances of recommender systems with regard to both individual users and groups. During the selection of the items for a group, the role of the corresponding relationships (e.g., position, dependency, and the strength of the social ties) is often more important than the individual preferences; however, the existing works do not sufficiently consider this important factor for group recommendations. We therefore propose a novel recommendation method that is based on a social affinity between the common histories of users. The proposed method consists of an intermovie similarity calculation that is based on weighted features for the generation of an initial social-affinity graph, and the subsequent computation of a user’s affinity to a group that is based on the graph. To apply the method for a service, we developed a “MyMovieHistory” application for the Facebook social media platform, and the synthetic dataset results of the experiment show that our proposed method can discover social affinities in an efficient manner.

GRSAT: A Novel Method on Group Recommendation by Social Affinity and Trustworthiness

ABSTRACT Existing group recommender systems generate a consensus function to aggregate individual preference into group preference. However, the systems encounter difficulty in gathering rating-scores and validating their reliability, since the aggregation strategy requires user rating-scores. To solve these problems, we propose Group Recommendation based on Social Affinity and Trustworthiness (GRSAT) based on social affinity and trustworthiness, which is obtained from the user’s watching-history and content features, without rating-score. Our experiment proves that GRSAT has outstanding performance for group recommendation compared with the other consensus functions, in terms of the number of the movies and users, on both biased and unbiased groups.

Serendipity-based storification: from lifelogging to storytelling

Storification is a theoretical technique which aims to construct the underlying relationships from discrete information for packaging them into a logical structure. In this paper, we focus on proposing the definition of serendipity-based storification in the personal history which is the combination of two-step processes: i) discovering hidden stories in the personal history and ii) representing stories using visualization techniques for easily grasping the information. MyMovieHistory Hong & Jung (Cybern Syst 46 (1-2), 69–83 ??) is used as the case study to demonstrate the effect of storification through detecting and presenting patterns in real personal data. The results can be utilized in helping people easily memorize and comprehend their histories. Moreover, additional benefits (e.g., reminding the pass, predicting the future, and communicating who you are) can be gained through the use of storification in the personal history.

Multi-Sided recommendation based on social tensor factorization

Abstract Tensor factorization has been applied in recommender systems to discover latent factors between multidimensional data such as time, place, and social context. However, tensor-based recommender systems still encounter with several problems such as sparsity, cold-start, and so on. In this paper, we introduce the new model social tensor to propose a tensor-based recommendation with a social relationship to deal with the existing problems. In addition, an adaptive method is presented to adjust the range of the social network for an active user. To evaluate our method, we conducted several experiments in the movie domain. The results indicate the ability of our method to improve the recommendation performance, even in the case of a new user. Particularly, the proposed method conducts the regeneration and factorization of the tensor in real time. Furthermore, our approach recommends not only a single item, but also the multi-factors for the item such as social, temporal, and spatial contexts.

Adaptive Complex Event Processing Based on Collaborative Rule Mining Engine

Complex Event Processing (CEP) detects complex events or patterns of event sequences based on a set of rules defined by a domain expert. However, it lowers the reliability of a system as the set of rules defined by an expert changes along with dynamic changes in the domain environment. A human error made by an expert is another factor that may undermine the reliability of the system. In an effort to address such problems, this study introduces Collaborative Rule Mining Engine (CRME) designed to automatically mine rules based on the history of decisions made by a domain expert by adopting a collaborative filtering approach, which is effective in mimicking and predicting human decision-making in an environment where there are sufficient data or information to do so. Furthermore, this study suggests an adaptive CEP technique, which does not hamper the reliability since it prevents potential errors caused by mistakes of domain experts and adapts to changes in the domain environment on its own as it is linked to the system proposed by Bharagavi [10]. In a bid to verify this technique, an automated stocks trading system will be established and its performance will be measured using the rate of return.

Visualizing Multidimensional Lifelogging Data: A Case Study on MyMovieHistory Project

ABSTRACT A large amount of time-series data has been frequently used to extract the useful patterns and trends and to visualize them for better understanding. This work is focusing on visualizing personal lifelogging data for tracking back to personal histories. Thereby, we present several similarity measures between multidimensional data at two different time points. For human evaluation, the method has been applied to MyMovieHistory (which a social recommendation system by storing personal movie logs) and tested with many users. Experimental results shown that the proposed visualization method and interfaces can help to understand user history.

Towards Affective Lifelogging with Information Fusion

Recently, most of context-aware services are trying to exploit the emotional contexts of the target users. The aim of this conceptual paper is to discuss affective lifelogging framework which can recognize the emotions by integrating multimodal information from multiple sources. Moreover, we will mention the open problems on affective lifelogging.

Understanding Your History: Multi-dimensional Data Stream Visualization of Personal Lifelogging Data

A large amount of time-series data has been frequently used to extract the useful patterns and trends and to visualize them for better understanding. This work is focusing on visualizing personal lifelogging data for tracking back to personal histories. Thereby, we present several similarity measures between multi-dimensional data at two different time points. For human evaluation, the method has been applied to MyMovieHistory (which a social recommendation system by storing personal movie logs) and tested with a number of users.

Social Affinity-Based Group Recommender System

Information collected from the social network is recently used to improve a performance of recommender systems to an individual user or a group. During selecting the items among the group members, the relationships (e.g., position, dependency, and the strength of the social ties) often has an important role than the individual preference in the group. Hence, we propose a novel recommendation method based on social affinity between two users. This recommendation method consists of (i) the similarity calculation between movies based on weighted feature, (ii) the generation of initial affinity network graph, and (iii) the computation of user’s affinity to group based on the graph. Experimental results on synthetic dataset show that our proposed method can discover social affinities efficiently.