• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.3
• /
• pp.123-128
• /
• 2019

This paper proposes the stock price prediction based on the artificial intelligence, where the model with recurrent convolution neural network (RCNN) layers is adopted. In the motivation of this prediction, long short-term memory model (LSTM)-based neural network can make the output of the time series prediction. On the other hand, the convolution neural network provides the data filtering, averaging, and augmentation. By combining the advantages mentioned above, the proposed technique predicts the estimated stock price of next day. In addition, in order to emphasize the recent time series, a custom weighted loss function is adopted. Moreover, stock data related to the stock price index are adopted to consider the market trends. In the experiments, the proposed stock price prediction reduces the test error by 3.19%, which is over other techniques by about 19%.

• KIISE Transactions on Computing Practices
• /
• v.20 no.11
• /
• pp.610-614
• /
• 2014

Like many types of scientific data, results from simulations of volcanic ash diffusion are of a clustered sparse matrix in the netCDF format. Since these data sets are large in size, they generate high storage and transmission costs. In this paper, we suggest a new method that reduces the size of the data of volcanic ash diffusion simulations by converting the multi-dimensional index to a single dimension and keeping only the starting point and length of the consecutive zeros. This method presents performance that is almost as good as that of ZIP format compression, but does not destroy the netCDF structure. The suggested method is expected to allow for storage space to be efficiently used by reducing both the data size and the network transmission time.

• Annual Conference of KIPS
• /
• 2002.04a
• /
• pp.11-14
• /
• 2002

분산 공간 데이터베이스 시스템에서 자주 수행되는 공간 죠인 질의는 공간 데이터의 특징인 대용량성과 복잡성으로 인하여 공간 연산 수행시 연간을 수행하는 서버의 CPU 및 디스크 I/O상의 과부하를 일으킨다. 본 논문은 이러한 분산 광간 데이터베이스 시스템에서 수행 비용이 많이 드는 원격 사이트간의 공간 죠인 질의를 병렬적이며 연쇄적으로 수행하는 기법을 제안한다. 본 기법은 공간 죠인 연산의 대상이 되는 릴레이션들을 공간 연산의 특성에 따라 순서화하고, 그 중 최하위의 죠인에 참여하는 릴레이션들 중 하나를 이등분 하는 방법으로 공간 죠인 연산을 분리한 추, 질의 수행에 참여하는 두 서버에게 죠인 연산을 분배한다. 각 서버는 분할된 공간 죠인 연산을 동시에 연쇄적으로 저리하고 결과를 병합하여 최종 죠인 결과를 생성한다. 본 기법은 릴레이션을 분할하여 죠인을 수행함으로써 공간 연산에 참여하는 객체의 수를 절반으로 줄이며 R-Tree 등의 공간 인덱스 탐색 횟수와 그 범위를 감소시킨다. 또한 연쇄적인 질의 처리로 죠인의 결과인 임시 릴레이션을 생성하지 않으므로 대용량의 데이터에 대한 복잡한 질의에 대해서도 제한 없이 수행한다.

• Journal of KIISE:Databases
• /
• v.28 no.3
• /
• pp.373-383
• /
• 2001

The distance join is a spatial join which finds data pairs in the order of distance when associating two spatial data sets. This paper proposes several methods to optimize the adaptive multi-stage distance join, presented in [1]. First, we optimize the sweeping index formula which is used for selecting sweeping axis during plane sweeping. Second, to improve the performance of a priority queue used for maintaining node pairs, we propose to use the maximum distance of a node pair as the second priority of the queue. Moreover, we compare trade-offs in estimating the cut-off distance between under uniformity assumption of data distribution and non-uniformity assumption. The experiments show that the proposed methods greatly improve the performance of the algorithm in CPU cost as well as in I/O cost.

• Annual Conference of KIPS
• /
• 2004.11a
• /
• pp.77-80
• /
• 2004

최근 XML은 전자상거래에서 의학, 국방, 법률 등의 전문분야에 이르기까지 많은 분야에서 활용되고 있으며, 데이터의 양 또한 방대해지고 있다. 따라서 대량의 XML 문서들을 효과적으로 저장하고 빠르게 검색할 수 있는 많은 인덱싱 기법들이 연구되고 있다. 최근의 인덱싱 기법들 중 Numbering Scheme 을 기반으로 한 인덱싱 기법들은 대부분의 검색에 우수한 성능을 보이나 하위노드의 수가 늘어나면 검색 오버헤드가 커질 수 있으며, 대량의 XML 문서의 추가 삽입 및 구조가 다른 XML 문서의 삽입시에 인덱스와 데이터 값의 재조정에 따른 많은 비용이 발생하게 된다. 이에 우리는 Numbering Scheme 을 기반으로 하지만, 각 노드별로 노드범위(Node-Range)와 Order-Array를 추가하여 검색성능을 향상시키고 대량의 XML 문서의 삽입 및 구조가 다른 XML 문서의 삽입시에 발생되는 문제를 해결하고자 한다.

• KIPS Transactions on Computer and Communication Systems
• /
• v.6 no.12
• /
• pp.487-496
• /
• 2017

In MySQL InnoDB, there are two ways of storing data. One is to create a separate tablespace for each table and store it separately. Another is to store all table and index information in a single system tablespace. You can use this information to recover deleted data from the record. However, in most of the current database forensic studies, the former is actively researched and its structure is analyzed, whereas the latter is not enough to be used for forensics. Both approaches must be analyzed in terms of database forensics because their storage structures are different from each other. In this paper, we propose a method for recovering deleted records in a method of storing records in IBDATA file, which is a single system tablespace. First, we analyze the IBDATA file to reveal its structure. And introduce delete record recovery algorithm which extended to an unallocated page area which was not considered in the past. In addition, we show that the recovery rate is improved up to 68% compared with the existing method through verification using real data by implementing the algorithm as a tool.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.6
• /
• pp.141-156
• /
• 2023

This study aims to improve understanding of freight movement, crucial for logistics facility investment and policy making. It addresses the limitations of traditional freight truck traffic data, aggregated only at city and county levels, by developing a new methodology. This method uses trip chain data for more detailed, facility-level analysis of freight truck movements. It employs DTG (Digital Tachograph) data to identify individual truck visit locations and creates H3 system-based polygons to represent these visits spatially. The study also involves an algorithm to dynamically determine the optimal spatial resolution of these polygons. Tested nationally, the approach resulted in polygons with 81.26% spatial fit and 14.8% error rate, offering insights into freight characteristics and enabling clustering based on traffic chain characteristics of freight trucks and visited facility types.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.24 no.4
• /
• pp.739-751
• /
• 2014

We are using our smartphone for our business as well as ours lives. Thus, user's privacy data and a company secret are stored at smartphone. By the way, the saved data on smartphone database can be exposed to a malicous attacker when a malicous app is installed in the smartphone or a user lose his/her smartphone because all data are stored as form of plaintext in the database. To prevent this disclosure of personal information, we need a database encryption method. However, if a database is encrypted, it causes of declining the performance. For example, when we search specific data in condition with encrypted database, we should decrypt all data stored in the database or search sequentially the data we want with accompanying overhead[1]. In this paper, we propose an efficient and searchable encryption method using variable length bloom filter under limited resource circumstances(e.g., a smartphone). We compare with existing searchable symmetric encryption. Also, we implemented the proposed method in android smartphone and evaluated the performance the proposed method. As a result through the implementation, We can confirm that our method has over a 50% improvement in the search speed compared to the simple search method about encrypted database and has over a 70% space saving compared to the method of fixed length bloom filter with the same false positive rate.

• The Journal of the Korea Contents Association
• /
• v.9 no.1
• /
• pp.23-29
• /
• 2009

This paper presents a capacity estimation of a storage system accumulating all data sensed during the lifetime of an individual human being. The calculation assumes modern data compression and data collection schemes based on wearable or implanted devices under ubiquitous environment. More than 76% of the storage area is found to be used for video data storage of common TV image quality. The remaining storage area is for data from other sensing organs including audio, taste, olfactory and tactual systems in addition to indexing information. Total storage area of around 600 tera bytes is needed to cover 100 years of human life including his fetal period.

• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.9
• /
• pp.377-384
• /
• 2015

In this paper we propose a visualization tool for distortion-free time-series matching. Supporting distortion-free is a very important factor in time-series matching to get more accurate matching results. In this paper, we visualize the result of time-series matching, which removes various time-series distortions such as noise, offset translation, amplitude scaling, and linear trend by using moving average, normalization, linear detrending transformations, respectively. The proposed visualization tool works as a client-server model. The client sends a user-selected time-series, of which distortions are removed, to the server and visualizes the matching results. The server efficiently performs the distortion-free time-series matching on the multi-dimensional R*-tree index. By visualizing the matching result as five different charts, we can more easily and more intuitively understand the matching result.