• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.357-359
• /
• 2023

This paper proposes a hybrid recommendation system (RS) model that overcomes the limitations of traditional approaches such as data sparsity, cold start, and scalability by combining collaborative filtering and context-aware techniques. The objective of this model is to enhance the accuracy of recommendations and provide personalized suggestions by leveraging the strengths of collaborative filtering and incorporating user context features to capture their preferences and behavior more effectively. The approach utilizes a novel method that combines contextual attributes with the original user-item rating matrix of CF-based algorithms. Furthermore, we integrate k-mean++ clustering to group users with similar preferences and finally recommend items that have highly rated by other users in the same cluster. The process of partitioning is the use of the rating matrix into clusters based on contextual information offers several advantages. First, it bypasses of the computations over the entire data, reducing runtime and improving scalability. Second, the partitioned clusters hold similar ratings, which can produce greater impacts on each other, leading to more accurate recommendations and providing flexibility in the clustering process. keywords: Context-aware Recommendation, Collaborative Filtering, Kmean++ Clustering.

• Journal of the Korea Computer Industry Society
• /
• v.2 no.9
• /
• pp.1175-1182
• /
• 2001

One to one Marketing (a.k.a. database marketing or relationship marketing) is one of the many fields that will benefit from the electronic revolution and shifts in consumer sales and advertising. As a component of intelligent customer services on Internet storefront, this paper describes technology of providing personalized advertisement using the market basket analysis, a well-Known data mining technique. The underlining theories of recommendation techniques are statistics, data mining, artificial intelligence, and/or rule-based matching. In the rule-based approach for personalized recommendation, marketing rules for personalization are usually collected from marketing experts and are used to inference with customer's data. However, it is difficult to extract marketing rules from marketing experts, and also difficult to validate and to maintain the constructed Knowledge base. In this paper, using marketing basket analysis technique, marketing rules for cross sales are extracted, and are used to provide personalized advertisement selection when a customer visits in an Internet store.

• Journal of Korea Spatial Information System Society
• /
• v.6 no.1 s.11
• /
• pp.59-71
• /
• 2004

Nearest neighbor query retrieves nearest located target objects, and is very frequently used in mobile environment. In this paper we propose a novel neatest neighbor query processing technique that is able to retrieve nearest located target object from the user who is continuously moving with a direction. The proposed method retrieves objects using the direction property of moving object as well as euclidean distance to target object. The proposed method is applicable to traffic information system, travel information system, and location-based recommendation system which require retrieving nearest located object.

• The Journal of Korean Physical Therapy
• /
• v.18 no.6
• /
• pp.23-32
• /
• 2006

Purpose: The purpose of this study is to suggest new strategies available for the physical therapy curriculum development. Method: The curriculum of 4 Universities in 4 countries; America, Australia, Canada and Korea was compared to suggest new curriculum. Results: Overall, curriculum in Korea emphasized skill and technique areas and didn't showed many subjects to take foundation of subject for understanding principles. The experience in clinic is not enough to satisfy international recommendation. Conclusion: We suggest that a new curriculum should be based on the three part which are foundation, essential and selection subject, and extended clinical experience to essentially need to be physical therapy in the world-standard.

• Journal of KIISE
• /
• v.43 no.3
• /
• pp.289-295
• /
• 2016

Malware authors spread malware variants in order to evade detection. It's hard to detect malware variants using static analysis. Therefore dynamic analysis based on API call information is necessary. In this paper, we proposed a malware family recommendation method to assist malware analysts in classifying malware variants. Our proposed method extract API call information of malware families by dynamic analysis. Then the multiple sequence alignment technique was applied to the extracted API call information. A signature of each family was extracted from the alignment results. By the similarity of the extracted signatures, our proposed method recommends three family candidates for unknown malware. We also measured the accuracy of our proposed method in an experiment using real malware samples.

• Journal of Korea Multimedia Society
• /
• v.9 no.12
• /
• pp.1709-1718
• /
• 2006

Users are able to use the information and service more free than previous wire network due to development of wireless network and device. For this reason, various studies on ubiquitous networks have been conducted. Various contexts brought in this ubiquitous environment, have recognized user's action through sensors. This results in the provision of better services. Because services exist in various places in ubiquitous networks, the application has the time of services searching. In addition, user's context is very dynamic, so a method needs to be found to recommend services to user by context. Therefore, techniques for reducing the time of service and increasing accuracy of recommendation are being studied. But it is difficult to quickly and appropriately provide large numbers of services, because only basic context information is stored. For this reason, we suggest DUPS(Dimension User Profile System), which stores location, time, and frequency information of often used services. Because previous technique used to simple information for recommending service without predicting services which is going to use on future, we can provide better service, and improve accuracy over previous techniques.

• Journal of Korea Multimedia Society
• /
• v.22 no.11
• /
• pp.1313-1323
• /
• 2019

With the development of the Internet and the increase of smart phones, various services considering user convenience are increasing, so that users can check news in real time anytime and anywhere. However, online news is categorized by media and category, and it provides only a few related search terms, making it difficult to find related news related to keywords. In order to solve this problem, we propose a method to recommend related documents more accurately by applying Doc2Vec similarity to the specific keywords of news articles and weighting the title and contents of news articles. We collect news articles from Naver politics category by web crawling in Java environment, preprocess them, extract topics using LDA modeling, and find similarities using Doc2Vec. To supplement Doc2Vec, we apply TF-IDF to obtain TC(Title Contents) weights for the title and contents of news articles. Then we combine Doc2Vec similarity and TC weight to generate TC weight-similarity and evaluate the similarity between words using PMI technique to confirm the keyword association.

• The Journal of Society for e-Business Studies
• /
• v.20 no.2
• /
• pp.155-166
• /
• 2015

RWR (Random Walk with Restart) is frequently used by many graph-based ranking algorithms, but it does not consider a signed graph where edges may have negative weight values. In this paper, we apply the Balance Theory by F. Heider to RWR over a signed graph and propose a novel RWR, Balanced Random Walk (BRW). We apply the proposed technique into the domain of recommendation system, and show by experiments its effectiveness to filter out the items that users may dislike. In order to provide the reasonable performance of BRW in the domain, we modify the existing Top-k algorithm, BCA, and propose a new algorithm, Bicolor-BCA. The proposed algorithm yet requires employing a threshold. In the experiment, we show how threshold values affect both precision and performance of the algorithm.

• Journal of the Korea Computer Industry Society
• /
• v.3 no.2
• /
• pp.167-178
• /
• 2002

XML documents used for exchanging structured documents between heterogeneous distributed systems are spreading among B2B and the industrial world rapidly. This condition brings an object-oriented visualization tool for modeling XML documents. It is hard to apply for DTD - the rule for declaring document type, which is employed in XML, to various industrial field. Thus, W3C announced Recommendation for XML Schema - the rule for declaring new document type which is adaptable to XML and satisfy to user. Document Type which has high reusability and flexibility can be defined owing to that XML Schema is designed by utilization of Object-Oriented-Modelling technique(UML). This document is proposed the specification and an algorithm for mapping XML Schema to UML.

• Journal of Korea Multimedia Society
• /
• v.22 no.9
• /
• pp.1108-1121
• /
• 2019

As the quality and level of life rise, many people are doing search for various pieces of information about tourism. In addition, users prefer the search methods reflecting individual opinions such as SNS and blogs to the official websites of tourist destination. Many of previous studies focused on a recommendation system for tourist courses based on the GPS information and past travel records of users, but such a system was not capable of recommending the latest tourist trends. This study thus set out to collect and analyze the latest SNS data to recommend tourist destination of high interest among users. It also aimed to propose an altered TSP algorithm to recommend the optimal routes to the recommended destination within an area and a system to recommend the optimal tourist courses by applying the Elasticsearch engine. The altered TSP algorithm proposed in the study used the location information of users instead of Dijkstra's algorithm technique used in previous studies to select a certain tourist destination and allowed users to check the recommended courses for the entire tourist destination within an area, thus offering more diverse tourist destination recommendations than previous studies.