• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.249-263
• /
• 2023

Collaborative Filtering, a representative recommendation system methodology, consists of two approaches: neighbor methods and latent factor models. Among these, the latent factor model using matrix factorization decomposes the user-item interaction matrix into two lower-dimensional rectangular matrices, predicting the item's rating through the product of these matrices. Due to the factor vectors inferred from rating patterns capturing user and item characteristics, this method is superior in scalability, accuracy, and flexibility compared to neighbor-based methods. However, it has a fundamental drawback: the need to reflect the diversity of preferences of different individuals for items with no ratings. This limitation leads to repetitive and inaccurate recommendations. The Adaptive Deep Latent Factor Model (ADLFM) was developed to address this issue. This model adaptively learns the preferences for each item by using the item description, which provides a detailed summary and explanation of the item. ADLFM takes in item description as input, calculates latent vectors of the user and item, and presents a method that can reflect personal diversity using an attention score. However, due to the requirement of a dataset that includes item descriptions, the domain that can apply ADLFM is limited, resulting in generalization limitations. This study proposes a Generalized Adaptive Deep Latent Factor Recommendation Model, G-ADLFRM, to improve the limitations of ADLFM. Firstly, we use item ID, commonly used in recommendation systems, as input instead of the item description. Additionally, we apply improved deep learning model structures such as Self-Attention, Multi-head Attention, and Multi-Conv1D. We conducted experiments on various datasets with input and model structure changes. The results showed that when only the input was changed, MAE increased slightly compared to ADLFM due to accompanying information loss, resulting in decreased recommendation performance. However, the average learning speed per epoch significantly improved as the amount of information to be processed decreased. When both the input and the model structure were changed, the best-performing Multi-Conv1d structure showed similar performance to ADLFM, sufficiently counteracting the information loss caused by the input change. We conclude that G-ADLFRM is a new, lightweight, and generalizable model that maintains the performance of the existing ADLFM while enabling fast learning and inference.

• Journal of Intelligence and Information Systems
• /
• v.14 no.1
• /
• pp.1-17
• /
• 2008

Data mining is defined as the process of discovering meaningful and useful pattern in large volumes of data. In particular, finding associations rules between items in a database of customer transactions has become an important thing. Some data structures and algorithms had been proposed for storing meaningful information compressed from an original database to find frequent itemsets since Apriori algorithm. Though existing method find all association rules, we must have a lot of process to analyze association rules because there are too many rules. In this paper, we propose a new data structure, called a Frequent Pattern Network (FPN), which represents items as vertices and 2-itemsets as edges of the network. In order to utilize FPN, We constitute FPN using item's frequency. And then we use a clustering method to group the vertices on the network into clusters so that the intracluster similarity is maximized and the intercluster similarity is minimized. We generate association rules based on clusters. Our experiments showed accuracy of clustering items on the network using confidence, correlation and edge weight similarity methods. And We generated association rules using clusters and compare traditional and our method. From the results, the confidence similarity had a strong influence than others on the frequent pattern network. And FPN had a flexibility to minimum support value.

• 한국벤처창업학회:학술대회논문집
• /
• 2017.04a
• /
• pp.41-41
• /
• 2017

연구에서는 기존의 업종구분 없이 창업보육센터의 서비스와 입주기업의 성과와의 상관관계 분석과 관련된 연구들을 문화콘텐츠 분야의 기업 관점에서도 유의미한지 재 검증해보고, 또한 문화콘텐츠 기업의 특성상 아이템의 사업성이 입주기업의 성과에 유의미한 영향을 미치는지에 대해 검증하고자 한다. 그래서 문화콘텐츠 분야의 창업보육 시 필요한 서비스를 발굴하고 빠르게 변화하는 문화콘텐츠 시장의 성장에 따른 아이템의 사업성 평가에 따른 차별화된 창업보육 서비스를 통해 성과가 도출될 수 있도록 체계적이고 합리적인 문화콘텐츠 분야의 창업보육시스템 구축의 중요성에 대해서 논의하고자 한다. 실증분석을 위하여 설문조사는 경기도 문화콘텐츠분야 창업보육센터 입주기업을 대상으로 설문을 실시하였고, 설문 데이터의 실증분석 방법으로는 SmartPLS 2.0를 이용해 빈도분석(頻度分析), 신뢰도분석(信賴度分析), 그리고 구조모형을 통하여 가설검증을 실시하였다. 가설검증을 통해 분석한 결과를 다음과 같다. 첫째, 경기도 문화콘텐츠 분야 창업보육센터의 지원서비스 수준이 창업아이템의 사업타당성에 미치는 영향을 분석한 결과에서는, 공간 및 부대지원서비스, 경영지원서비스, 기술지원 서비스는 창업아이템의 사업타당성에는 유의한 영향을 미치지 않는 것으로 나타났으며, 인적지원 서비스 및 마케팅 지원 서비스는 유의미한 정(+)의 영향을 미치는 것을 확인 할 수 있었다. 둘째, 경기도 문화콘텐츠 분야 창업보육센터의 지원서비스 수준이 입주기업의 성과에 미치는 영향을 분석한 결과에서는, 공간 및 부대지원 서비스, 인적지원 서비스 및 마케팅지원 서비스는 입주기업의 재무적 성과에 유의미한 정(+)의 영향을 미치는 것을 확인 할 수 있었으나, 경영지원 서비스, 기술지원 서비스는 입주기업의 재무적 성과에 유의미한 영향을 미치지 않는 것으로 나타났다. 또한 공간 및 부대지원 서비스, 경영지원 서비스, 인적지원 서비스 및 마케팅지원 서비스는 입주기업의 비재무적 성과에 유의미한 정(+)의 영향을 미치는 것을 확인 할 수 있었으나. 기술지원 서비스는 입주기업의 비재무적 성과에 유의한 영향을 미치지 않는 것으로 나타났다. 셋째, 경기도 문화콘텐츠 분야 창업아이템의 사업타당성이 입주기업의 성과에 미치는 영향을 미치는 것을 분석한 결과 창업아이템의 사업타당성이 입주기업의 성과(재무적,비재무적) 성과에 모두 영향을 미치는 중요한 요소임을 확인 할 수 있었다.

• Journal of Broadcast Engineering
• /
• v.8 no.1
• /
• pp.80-90
• /
• 2003

Today, thanks to the spread of Internet and the development of various multimedia technologies, broadcasting environment is changing in a way to merging broadcasting and communication environment. In such a new environment, transparent delivery and interoperability between different technologies become an important issue. In ISO/IEC MPEG meeting, MPEG-21 has been established. In MPEG-21, it is discussed actively about big picture to build an infrastructure for the delivery and consumption of multimedia contents. In this paper, we have applied Digital Item in MPEG-21 to broadcasting contents. Broadcasting Digital Item is generated using MPEG-7 MDS for the description of resources as well as Digital Item Declaration, and Digital Item Identification. Furthermore, we develop Digital Item generation system and parsing system. In experiment, we verify the usefulness of Broadcasting Digital Item by generating an education video contents and consuming it at different user environments.

• Journal of Broadcast Engineering
• /
• v.8 no.4
• /
• pp.410-426
• /
• 2003

MPEG-21 defines a digital item as an atomic unit lot creation, delivery and consumption in order to provide an integrated multimedia framework in networked environments. It is expected that MPEG-21 standardization makes it Possible for users to universally access user's preferred contents in their own way they want. In order to achieve this goal, MPEG-21 has standardized the specifications for the Digital Item Declaration (DID). Digital Identification (DII), Rights Expression Language (REL), Right Data Dictionary (RDD) and Digital Item Adaptation (DIA), and is standardizing the specifications for the Digital Item Processing (DIP), Persistent Association Technology (PAT) and Intellectual Property Management and Protection (IPMP) tot transparent and secured usage of multimedia. In this paper, we design an MPEG-21 terminal architecture based one the MPEG-21 standard with DID, DIA and DIP, and implement with the MPEG-21 terminal. We make a video summarization service scenario in order to validate ow proposed MPEG-21 terminal for the feasibility to of DID, DIA and DIP. Then we present a series of experimental results that digital items are processed as a specific form after adaptation fit for the characteristics of MPEG-21 terminal and are consumed with interoperability based on a PC and a PDA platform. It is believed that this paper has n important significance in the sense that we, for the first time, implement an MPEG-21 terminal which allows for a video summarization service application in an interoperable way for digital item adaptation and processing nth experimental results.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.01a
• /
• pp.173-174
• /
• 2022

최근 확률형 아이템이 이용자들에게 투입 비용보다 높은 가치의 아이템을 획득할 수 있을 것이라는 막연한 기대감을 갖게 만들어, 게임 이용자의 사행성을 조장하고 과소비를 부추긴다는 우려가 제기되고 있다. 특히 복잡한 확률구조와 관련하여 허위로 표시하는 등 거짓 또는 과장 광고가 만연하여 게임 업계의 자율규제가 실효성을 잃고 게임제작업자가 허위로 확률을 고지하여 이용자들을 기만하는 사례 등이 사회적으로 문제가 되고 있다. 이에 국민의힘 하태경 의원이 일정 규모 이상 게임제작업자는 게임물이용자위원회를 두도록 하는 게임산업진흥에 관한 법률 일부개정(안)을 2021년 3월 24일 대표발의하였는데, 이 법안이 게임산업을 과도하게 규제한다는 비판이 제기되고 있는바, 해당 법안을 분석하여 입법취지에 부합하는 개선방안을 제시하고자 한다.

• Journal of Broadcast Engineering
• /
• v.8 no.4
• /
• pp.351-364
• /
• 2003

In this paper, we propose visual Impairment descriptions for MPEG-21 digital item adaptation. The proposed visual impairment descriptions include low vision impairment and color vision deficiency. It is symptom-based description so that the description is systematic enough to sufficiently describe user's any kind of visual impairment characteristics and easy enough to describe for any user with visual impairment. In this paper, we performed some experiments with the proposed description in MPEG-21 digital item adaptation. The experiments showed that the proposed description Is effective to adapt visual resources In digital item for users with visual impairment and to give enhanced visual accessibility to them.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.10
• /
• pp.2597-2604
• /
• 1996

본 논문에서는 주기억 데이터베이스 시스템에서의 효율적인 데이터 처리를 위하여 T*-트리라는 새로운 색인구조를 제시한다. T*-트리 색인구조는 기존의 디스크를 기반으로 하는 색인기법과 달리 모든 데이터가 주기억장치에 적재되어 있는 시스템에서 보다 빠른 데이터 접근과 메모리 공간의 효율적인 사용을 위해 주기억 데이터베이스 시스템에서 주로 사용되고 있는 T*-트리색인구조의 장점은 그대로 계승하면서 단점을 보완한 인덱스 구조이다. 본 논문에서 제시하는 T*-트리는 데이터 아이템에 대한 검색과 저장공간의 활용면에서는 T*-트리와 대동소이한 성능을 가지고 있으나, 범위 질의에서와 데이터 아이템에 대한 검색과 삭제시 중간노드에서의 노드간의 순회경로를 줄임으로써 보다 향상된 성능을 보여준다. 또한 T*-트리와 스레드 이진트리를 조합하는 경우에는 순회경로가 다소 단축되지만 중간노드에서 자신보다 높은 레벨의 후속 노드로의 순회는 기존의 인오더 트리 순회에 의존하지만, T*-트리에서는 후위포인터를 이용하므로 직접순회가 가능하게 된다. 본 논문에서는 제안된 T*-트리의 구조와 T*-트리의 검색, 삽입 및 삭제 연산을 위한 알고리즘을 설명한 후, 기존의 T*-트리와 성능분석을 실시하고 그 결과를 제시한다. 성능 분석결과 T*-트리는 데이터 검색의 경우 기존의 T*-트리와 거의 동일한 성능을 보였으며, 삽입과 식제등 색인구조의 변경시는 약간의 성능향상을 보였으나, 범위질의와 순차질의에서는 매우 향상된 성능을 나타냈다.

• Journal of Internet Computing and Services
• /
• v.14 no.6
• /
• pp.1-8
• /
• 2013

Data mining techniques are used to find important and meaningful information from huge databases, and pattern mining is one of the significant data mining techniques. Pattern mining is a method of discovering useful patterns from the huge databases. Frequent pattern mining which is one of the pattern mining extracts patterns having higher frequencies than a minimum support threshold from databases, and the patterns are called frequent patterns. Traditional frequent pattern mining is based on a single minimum support threshold for the whole database to perform mining frequent patterns. This single support model implicitly supposes that all of the items in the database have the same nature. In real world applications, however, each item in databases can have relative characteristics, and thus an appropriate pattern mining technique which reflects the characteristics is required. In the framework of frequent pattern mining, where the natures of items are not considered, it needs to set the single minimum support threshold to a too low value for mining patterns containing rare items. It leads to too many patterns including meaningless items though. In contrast, we cannot mine any pattern if a too high threshold is used. This dilemma is called the rare item problem. To solve this problem, the initial researches proposed approximate approaches which split data into several groups according to item frequencies or group related rare items. However, these methods cannot find all of the frequent patterns including rare frequent patterns due to being based on approximate techniques. Hence, pattern mining model with multiple minimum supports is proposed in order to solve the rare item problem. In the model, each item has a corresponding minimum support threshold, called MIS (Minimum Item Support), and it is calculated based on item frequencies in databases. The multiple minimum supports model finds all of the rare frequent patterns without generating meaningless patterns and losing significant patterns by applying the MIS. Meanwhile, candidate patterns are extracted during a process of mining frequent patterns, and the only single minimum support is compared with frequencies of the candidate patterns in the single minimum support model. Therefore, the characteristics of items consist of the candidate patterns are not reflected. In addition, the rare item problem occurs in the model. In order to address this issue in the multiple minimum supports model, the minimum MIS value among all of the values of items in a candidate pattern is used as a minimum support threshold with respect to the candidate pattern for considering its characteristics. For efficiently mining frequent patterns including rare frequent patterns by adopting the above concept, tree based algorithms of the multiple minimum supports model sort items in a tree according to MIS descending order in contrast to those of the single minimum support model, where the items are ordered in frequency descending order. In this paper, we study the characteristics of the frequent pattern mining based on multiple minimum supports and conduct performance evaluation with a general frequent pattern mining algorithm in terms of runtime, memory usage, and scalability. Experimental results show that the multiple minimum supports based algorithm outperforms the single minimum support based one and demands more memory usage for MIS information. Moreover, the compared algorithms have a good scalability in the results.

• The KIPS Transactions:PartB
• /
• v.12B no.2 s.98
• /
• pp.137-142
• /
• 2005

Since wireless communication cost is relatively expensive, the mobile games using cellular phones or PDAs are mostly played after being downloaded unlike more updated types of phone-to-phone or online games. In particular, the upbringing games, in which garners foster virtual companion animals or vegetables as their avata, have been spotlighted. It is essential to supply variety of items for upbringing avatas. However, due to diverse tastes for avatas and limited storage of mobile terminals, game manias must download their desired items from homepages. In addition, game developers must inform users through SMS messages whenever a new item is created. To do so, they must link WEB server and WAP server to Call Back URL or ARS. This paper proposed a linkage method suitable for JAVA-based mobile phone operating system. The proposed method will consequently increase life cycle of a game and reinforce profitability.