• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.4
• /
• pp.721-727
• /
• 2016

In this paper, in order to resolve the problem of reduction for D2D (device to device) advertisement dissemination efficiency of conventional dissemination algorithms, we here propose several clustering algorithms (modified single linkage algorithm (MSL), K-means algorithm, and expectation maximization algorithm with Gaussian mixture model (EM)) based advertisement dissemination algorithms to improve advertisement dissemination efficiency in D2D communication networks. Target areas are clustered in several target groups by the proposed clustering algorithms. Then, D2D advertisements are consecutively distributed by using a routing algorithm based on the geographical distribution of the target areas and a relay selection algorithm based on the distance between D2D sender and D2D receiver. Via intensive MATLAB simulations, we analyze the performance excellency of the proposed algorithms with respect to maximum number of relay transmissions and D2D user density ratio in a target area and a non-target area.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.05a
• /
• pp.40-42
• /
• 2015

Our D2D (Device-to-Device) advertisement dissemination algorithm based on the convergence of a D2D communication and a social commerce service is able to efficiently deliver advertisement messages of small business owners. We here assume high user-density areas as target-areas, and these target areas can form several target-groups based on geographical proximity. Also, the isolated target-areas which are not included in the target-groups may have the opportunities for forming a new target-group in consideration of the maximum distance with AP. In this paper, we evaluate performances with respect to the total number of successfully received users and the average number of relay users in accordance with variation in the maximum distance for isolated target-areas.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.10
• /
• pp.2403-2408
• /
• 2014

We designate multiple target areas for the advertisement disseminations in order to resolve the problem of advertisement transmission efficiency degradation due to overlapped device-to-device (D2D) transmissions and unnecessary advertisement transmissions. We here propose an efficient advertisement dissemination algorithm based on pre-selected target areas considering user density. In our proposed algorithm, relay nodes are gradually selected according to the locations of the target areas. We mathematically analyze D2D advertisement coverages of our proposed algorithm, and compare the various simulation results of the proposed algorithm with those of the conventional algorithm according to the several simulation scenarios via intensive simulations.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.2
• /
• pp.287-292
• /
• 2015

In this paper, we evaluated the performance of D2D advertisement dissemination algorithms with maximum-distance and transmission-efficiency based relay selections with respect to the total number of successfully received users and transmission efficiency. To assume more practical environment, we took into account pre-defined target-areas based on the information of user density and the limit for the maximum number of relay users. Through the simulations we compared the performance results of both D2D advertisement dissemination algorithms with maximum-distance and transmission-efficiency based relay selections according to increment of the number of sectors. And then, we analysed the superiority of algorithm with transmission-efficiency based relay selections more than maximum-distance based relay selections.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.42 no.5
• /
• pp.992-998
• /
• 2017

For local advertising based on D2D communications, sources want advertisement messages to be diffused to unspecified users as many as possible. It is one of challenging issues to select target-areas for advertising if users are uniformly distributed. In this paper, we propose D2D based advertisement dissemination algorithm using user clustering with expectation-maximization. The user distribution of each cluster can be estimated by principal components (PCs) obtained from each cluster. That is, PCs enable the target-areas and routing paths to be properly determined according to the user distribution. Consequently, advertisement messages are able to be disseminated to many users. We evaluate performances of our proposed algorithm with respect to coverage probability and average reception number per user.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.07a
• /
• pp.279-280
• /
• 2020

최근 소셜 미디어로 이름을 알린 이색 카페와 맛집을 찾아다니는 문화가 확산되는 추세이다. 블로그 포털 검색을 통해 찾아본 맛집은 광고성 게시물이 많아서 신뢰도가 떨어지고, 맛집 관련 게시물 수가 많아서 모든 게시물들을 수동으로 읽기는 불가능하다. 본 논문에서는 사용자들이 선호해서 자발적으로 공유하는 신뢰도 높은 인스타그램의 맛집 포스트 데이터를 이용하여 아이템 기반의 협업 필터링(Item-based Collaborative Filtering) 기법을 통해 사용자의 취향에 맞고 선호할 만한 맛집을 자동으로 추천해주는 알고리즘 및 시스템을 소개한다.

• Annual Conference on Human and Language Technology
• /
• 2017.10a
• /
• pp.294-297
• /
• 2017

SNS와 스마트기기의 발전으로 온라인을 통한 뉴스 배포가 용이해지면서 악의적으로 조작된 뉴스가 급속도로 생성되어 확산되고 있다. 뉴스 조작은 다양한 형태로 이루어지는데, 이 중에서 정상적인 기사 내에 광고나 낚시성 내용을 포함시켜 독자가 의도하지 않은 정보에 노출되게 하는 형태는 독자가 해당 내용을 진짜 뉴스로 받아들이기 쉽다. 본 논문에서는 뉴스 기사 내에 포함된 문단 중에서 부적합한 문단이 포함되었는지를 판정하기 위한 방법을 제안한다. 제안하는 방식에서는 자연어 처리에 유용한 Convolutional Neural Network(CNN)모델 중 Word2Vec과 tf-idf 알고리즘, 로지스틱 회귀를 함께 이용하여 뉴스 부적합 문단을 검출한다. 본 시스템에서는 로지스틱 회귀를 이용하여 문단의 카테고리를 분류하여 본문의 카테고리 분포도를 계산하고 Word2Vec을 이용하여 문단간의 유사도를 계산한 결과에 가중치를 부여하여 부적합 문단을 검출한다.

• 한국어정보학회:학술대회논문집
• /
• 2017.10a
• /
• pp.294-297
• /
• 2017

SNS와 스마트기기의 발전으로 온라인을 통한 뉴스 배포가 용이해지면서 악의적으로 조작된 뉴스가 급속도로 생성되어 확산되고 있다. 뉴스 조작은 다양한 형태로 이루어지는데, 이 중에서 정상적인 기사 내에 광고나 낚시성 내용을 포함시켜 독자가 의도하지 않은 정보에 노출되게 하는 형태는 독자가 해당 내용을 진짜 뉴스로 받아들이기 쉽다. 본 논문에서는 뉴스 기사 내에 포함된 문단 중에서 부적합한 문단이 포함 되었는지를 판정하기 위한 방법을 제안한다. 제안하는 방식에서는 자연어 처리에 유용한 Convolutional Neural Network(CNN)모델 중 Word2Vec과 tf-idf 알고리즘, 로지스틱 회귀를 함께 이용하여 뉴스 부적합 문단을 검출한다. 본 시스템에서는 로지스틱 회귀를 이용하여 문단의 카테고리를 분류하여 본문의 카테고리 분포도를 계산하고 Word2Vec을 이용하여 문단간의 유사도를 계산한 결과에 가중치를 부여하여 부적합 문단을 검출한다.

• Journal of Korea Multimedia Society
• /
• v.10 no.2
• /
• pp.269-277
• /
• 2007

Network clustering has been proposed to provide that sensor nodes minimize energy and maximize a network lifetime by configuring clusters, Although dynamic clustering brings extra overhead like as head changing, head advertisement, it may diminish the gain in energy consumption to report attribute tasks by using cluster heads. Therefore, this paper proposes a new routing algorithm which configures cluster to reduce the number of messages when establishing paths and reports to the sink by way of cluster heads when responding sens ing tasks. All sensor nodes only broadcast bitmap once and maintain a bitmap table expressed by bits, allowing them to reduce node energy and to prolong the network lifetime. After broadcasting, each node only updates the bitmap without propagation when the adjacent nodes broad cast same query messages, This mechanism makes nodes to have abundant paths. By modifying the query which requests sensing tasks, the size of cluster is designed dynamically, We try to divide cluster by considering the number of nodes. Then, all nodes in a certain cluster must report to the sub- sink node, The proposed routing protocol finds easily an appropriate path to report tasks and reduces the number of required messages for the routing establishment, which sensor nodes minimize energy and maximize a network lifetime.