• The Journal of Society for e-Business Studies
• /
• v.26 no.3
• /
• pp.67-80
• /
• 2021

The node embedding technique for learning graph representation plays an important role in obtaining good quality results in graph mining. Until now, representative node embedding techniques have been studied for homogeneous graphs, and thus it is difficult to learn knowledge graphs with unique meanings for each edge. To resolve this problem, the conventional Triple2Vec technique builds an embedding model by learning a triple graph having a node pair and an edge of the knowledge graph as one node. However, the Triple2 Vec embedding model has limitations in improving performance because it calculates the relationship between triple nodes as a simple measure. Therefore, this paper proposes a feature extraction technique based on a graph convolutional neural network to improve the Triple2Vec embedding model. The proposed method extracts the neighborliness vector of the triple graph and learns the relationship between neighboring nodes for each node in the triple graph. We proves that the embedding model applying the proposed method is superior to the existing Triple2Vec model through category classification experiments using DBLP, DBpedia, and IMDB datasets.

• The Journal of the Korea Contents Association
• /
• v.18 no.6
• /
• pp.217-234
• /
• 2018

In this study, big data analysis was conducted for domestic and international sports goods brands. Text Mining, TF-IDF, Opinion Mining, interestity graph were conducted through the social matrix program Textom and the fashion data analysis platform MISP. In order to examine the recent recognition of sports brands, the period of study is limited to 1 year from January 1, 2017 to December 31, 2017. As a result of analysis, first, we could confirm the products representing each brand. Second, I could confirm the marketing that represents each brand. Third, the common words extracted from each brand were identified. Fourth, the emotions of positive and negative of each brand were confirmed.

• Industrial Engineering and Management Systems
• /
• v.10 no.4
• /
• pp.238-246
• /
• 2011

Given a directed graph, we can determine how the user's preference moves from one product item to another. In this graph called "preference transition network", each node represents the product item while its edge pointing to the other nodes represents the transition of user's preference. However, with the large number of items make the network become more complex, unclear and difficult to be interpreted. In order to address this problem, this paper proposes a visualization technique in preference transition analysis based on recency and frequency. By adapting these two elements, the semantic meaning of each item and its transition can be clearly identified by its different types of node size, color and edge style. The experiment in a sales data has shown the results of the proposed approach.

• Genomics & Informatics
• /
• v.15 no.1
• /
• pp.19-27
• /
• 2017

Understanding complex relationships among heterogeneous biological data is one of the fundamental goals in biology. In most cases, diverse biological data are stored in relational databases, such as MySQL and Oracle, which store data in multiple tables and then infer relationships by multiple-join statements. Recently, a new type of database, called the graph-based database, was developed to natively represent various kinds of complex relationships, and it is widely used among computer science communities and IT industries. Here, we demonstrate the feasibility of using a graph-based database for complex biological relationships by comparing the performance between MySQL and Neo4j, one of the most widely used graph databases. We collected various biological data (protein-protein interaction, drug-target, gene-disease, etc.) from several existing sources, removed duplicate and redundant data, and finally constructed a graph database containing 114,550 nodes and 82,674,321 relationships. When we tested the query execution performance of MySQL versus Neo4j, we found that Neo4j outperformed MySQL in all cases. While Neo4j exhibited a very fast response for various queries, MySQL exhibited latent or unfinished responses for complex queries with multiple-join statements. These results show that using graph-based databases, such as Neo4j, is an efficient way to store complex biological relationships. Moreover, querying a graph database in diverse ways has the potential to reveal novel relationships among heterogeneous biological data.

• KIISE Transactions on Computing Practices
• /
• v.21 no.7
• /
• pp.482-487
• /
• 2015

This paper presents the results and the process of extracting correlations between events of prescriptions and examinations using graph-modeling and node centrality measures on a medical dataset of 11,938 patients with diabetes mellitus. As the data is stored in relational form, RDB2Graph framework was used to construct effective graph models from the data. Personalized PageRank was applied to analyze correlation between prescriptions and examinations of the patients. Two graph models were constructed: one that models medical events by each patient and another that considers the time gap between medical events. The results of the correlation analysis confirm current medical knowledge. The paper demonstrates some of the note-worthy findings to show the effectiveness of the method used in the current analysis.

• The Journal of the Korea Contents Association
• /
• v.22 no.1
• /
• pp.35-46
• /
• 2022

Recently, with the development of network technologies, as IoT and social network service applications have been actively used, a lot of graph stream data is being generated. In this paper, we propose a graph compression scheme that considers the stream graph environment by applying graph mining to the existing compression technique, which has been focused on compression rate and runtime. In this paper, we proposed Incremental frequent pattern based compression technique for graph streams. Since the proposed scheme keeps only the latest reference patterns, it increases the storage utilization and improves the query processing time. In order to show the superiority of the proposed scheme, various performance evaluations are performed in terms of compression rate and processing time compared to the existing method. The proposed scheme is faster than existing similar scheme when the number of duplicated data is large.

• Journal of Digital Contents Society
• /
• v.19 no.3
• /
• pp.471-477
• /
• 2018

As time changes, users change their interest. In this paper, we propose a method to provide suitable service for users by dynamically weighting service interests in the context of age, timing, and seasonal changes in ubiquitous environment. Based on the service history data presented to users according to the age or season, we also offer useful services by continuously adding the most recent service rules to reflect the changing of service interest. To do this, a set of services is considered as a transaction and each service is considered as an item in a transaction. And also we represent the association of services in a graph and extract frequent service items that refer to the latest information services for users.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.8
• /
• pp.2101-2123
• /
• 2023

Recent studies have shown that the neural network-based binary code similarity detection technology performs well in vulnerability mining, plagiarism detection, and malicious code analysis. However, existing cross-architecture methods still suffer from insufficient feature characterization and low discrimination accuracy. To address these issues, this paper proposes a cross-architecture binary function similarity detection method based on composite feature model (SDCFM). Firstly, the binary function is converted into vector representation according to the proposed composite feature model, which is composed of instruction statistical features, control flow graph structural features, and application program interface calling behavioral features. Then, the composite features are embedded by the proposed hierarchical embedding network based on a graph neural network. In which, the block-level features and the function-level features are processed separately and finally fused into the embedding. In addition, to make the trained model more accurate and stable, our method utilizes the embeddings of predecessor nodes to modify the node embedding in the iterative updating process of the graph neural network. To assess the effectiveness of composite feature model, we contrast SDCFM with the state of art method on benchmark datasets. The experimental results show that SDCFM has good performance both on the area under the curve in the binary function similarity detection task and the vulnerable candidate function ranking in vulnerability search task.

• Journal of KIISE:Computer Systems and Theory
• /
• v.33 no.5
• /
• pp.267-281
• /
• 2006

PPI(Protein-Protein Interaction) data has information about the organism has maintained a life with some kind of mechanism. So, it is used in study about cure research back, cause of disease, and new medicine development. This PPI data has been increased by geometric progression because high throughput methods are developed such as Yeast-two-hybrid, Mass spectrometry, and Correlated mRNA expression. So, it is impossible that a person directly manage and analyze PPI data. Fortunately, PPI data is able to abstract the graph which has proteins as nodes, interactions as edges. Consequently, Graph theory plentifully researched from the computer science until now is able to be applied to PPI data successfully. In this paper, we introduce Proteinca(PROTEin INteraction CAbaret) workbench system for easily managing, analyzing and visualizing PPI data. Proteinca assists the user understand PPI data intuitively as visualizing a PPI data in graph and provide various analytical function on graph theory. And Protenica provides a simplified visualization with gravity-rule.

• The KIPS Transactions:PartD
• /
• v.10D no.2
• /
• pp.261-266
• /
• 2003

A protein, which is a linear polymer of amino acids, is one of the most important bio-molecules composing biological structures and regulating bio-chemical reactions. Since the characteristics and functions of proteins are determined by their amino acid sequences in principle, protein sequence determination is the starting point of protein function study. This paper proposes a protein sequence prediction method based on data mining techniques, which can overcome the limitation of previous bio-chemical sequencing methods. After applying multiple proteases to acquire overlapped protein fragments, we can identify candidate fragment sequences by comparing fragment mass values with peptide databases. We propose a method to construct multi-partite graph and search maximal paths to determine the protein sequence by assembling proper candidate sequences. In addition, experimental results based on the SWISS-PROT database showing the validity of the proposed method is presented.