• Journal of KIISE:Databases
• /
• v.33 no.6
• /
• pp.631-643
• /
• 2006

Moving objects in a mobile environment to change their position based on the change of time require a query with their position as a basis. Efficient Regional Decision for Continuous Skyline Queries requires objectively pre-calculating the OSR(Optimal Skyline Region) regardless of the speed and direction of the moving objects. It proposes techniques to reduce the frequency of continuous queries by choosing a VCircle(Validity Circle) as safe location which has radius of the distance to the closest region with position on the moving object at center. But, a VCircle's area varies based on the Moving object's position from first marked time of continuous query. Therefore, the frequency of its continuous query is variable and also when the object moves inside of OSR, query can re-occur frequently In this paper, we suggest a technique of selecting an IVCircle(Interior Validity Circle) in a Skyline Region as the static Safe Region using the characteristics of the OSR. An Interior IVCircle can be calculated in advance when the OSR is decided. Our experiment shows that the frequency of using IVcircle as safe region reduced than that of using VCircle as safe region by 52.55%.

• Journal of KIISE:Databases
• /
• v.35 no.2
• /
• pp.182-191
• /
• 2008

Most of previous works for skyline queries have focused only on static attributes of target objects. With the advance in mobile applications, however, the need of continuous skyline queries for moving objects has been increasing. Even though several techniques to process continuous skyline queries have been proposed recently, they cannot process subspace queries, which use only the subset of attribute dimensions. Therefore it is not feasible to utilize those methods for mobile applications which must consider moving objects and subspaces simultaneously. In this paper, we propose a dominant object-based pruning method to compute subspace skyline of moving objects efficiently at query time and present the experimental results to show the effectiveness of the proposed method.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.12
• /
• pp.938-942
• /
• 2009

In this paper, we propose a novel filtering method for processing continuous skyline queries on wireless sensor network environments. The existing filtering methods use the filter based on router paths. However, because these filters are applied not to a whole area but to a partial area, these methods send almost data of sensor nodes to transmit to the base station and have no sufficient effect in terms of energy efficiency. Therefore, we propose an efficient method to dramatically reduce the transmission data of sensors through applying a low-cost and effective filter to all sensor nodes. The proposed effective filter is generated by using characteristics such as the data locality and the clustering of sensors. An extensive performance study verifies the merits of our new method.

• KIPS Transactions on Computer and Communication Systems
• /
• v.10 no.8
• /
• pp.231-234
• /
• 2021

Various problems arise as applications using locations increase. In order to solve this problem, related works are being conducted to protect the location of users. A fundamental reason for this problem is that users must provide their location information to the service provider (server) to receive the service. To improve these problems, there are works such as generating cloaking regions or generating dummies around them. However, if a user periodically asks the server for queries, the user's trajectory may be exposed by time zone. To improve this problem, in this paper, we propose a k-Anonymity Trajectory (k-ATY) technique that can improve the exposure probability of the trajectory even if the user requests continuous queries. Experimental results demonstrated the superiority of the proposed technique.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.4
• /
• pp.289-293
• /
• 2009

In sensor networks, many methods have been proposed to process in-network aggregation effectively. Contrary to normal aggregation queries, skyline query processing that compare multi-dimension data for producing result is very hard. It is important to filter unnecessary data for energy-efficient skyline query processing. Existing approach like MFTAC restricts unnecessary data transitions by deploying filters to whole sensors. However, network lifetime is reduced by energy consumption for filters transmission. In this paper, we propose a lazy filtering-based skyline query processing algorithm of in-network for reducing energy consumption by filters transmission. The proposed algorithm creates the skyline filter table (SFT) in the data gathering process which sends from sensor nodes to the base station and filters out unnecessary transmissions using it. The experimental results show that the proposed algorithm reduces false positive by 53% and improves network lifetime by 44% on average over MFTAC.

• Journal of KIISE:Databases
• /
• v.34 no.2
• /
• pp.166-178
• /
• 2007

The ALE specification of EPCglobal is leading the development of RFID standards, includes the Event Cycle Specification (ECSpec) describing how long a cycle is, how to filter RFID tag data and which reader is interested in. The ECSpec is a specification for filtering and collecting RFID tag data. It is registered to a middleware for long time and is evaluated to return results satisfying the requirements included in it. Thus, it is quite similar to the continuous query. It can be transformed into a continuous query as its predicate in WHERE clause is characterized by the long interval. Long intervals cause problems deteriorating insertion and search performance of existing query indices. In this paper, we propose a TLC-index as a new query index structure for long interval data. The TLC-index has hybrid structure that uses the cell construct of CQI-index with the virtual construct of VCR-index for partitioning long intervals. The TLC-index can reduce the storage cost and improve the insertion performance through decomposing long intervals into one or more cell constructs that have long size. It can also improve the search performance through decomposing short intervals into one or more virtual constructs that have short size enough to fit into those intervals.

• Journal of KIISE
• /
• v.43 no.5
• /
• pp.590-595
• /
• 2016

In this study, we propose an efficient method for monitoring continuous top-k queries. In contrast to the conventional top-k queries, the presented top-k query considers both spatial and non-spatial attributes. We proposed a novel main-memory based grid access method, called Bit-Vector Grid Index (BVGI). The proposed method quickly identifies whether the moving objects are included in some of the grid cell by encoding a non-spatial attribute value of the moving object to bit-vector. Experimental simulations demonstrate that the proposed method is several times faster than the previous method and uses considerably less memory.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.396-401
• /
• 2021

초거대 언어모델를 활용한 퓨샷(few shot) 학습법은 여러 자연어 처리 문제에서 좋은 성능을 보였다. 하지만 데이터를 활용한 추가 학습으로 문제를 추론하는 것이 아니라, 이산적인 공간에서 퓨샷 구성을 통해 문제를 정의하는 방식은 성능 향상에 한계가 존재한다. 이를 해결하기 위해 초거대 언어모델의 모수 전체가 아닌 일부를 추가 학습하거나 다른 신경망을 덧붙여 연속적인 공간에서 추론하는 P-tuning과 같은 데이터 기반 추가 학습 방법들이 등장하였다. 본 논문에서는 문맥에 따른 질의 정규화 문제를 대화형 음성 검색 서비스에 맞게 직접 정의하였고, 초거대 언어모델을 P-tuning으로 추가 학습한 경우 퓨샷 학습법 대비 정확도가 상승함을 보였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.11b
• /
• pp.655-658
• /
• 2003

기존의 데이터베이스 응용 프로그램을 작성할 경우에 질의 언어와 개발 언어간의 이중 언어 정의 문제(impedance mismatch)가 발생한다. 이를 해결하기 위한 한가지 방법으로 본 논문에서는 Haskell의 리스트 컴프리헨션(list comprehension)을 이용한 데이터베이스 접근을 제안한다. 이 방식을 이용할 경우 문법 체계가 명확해 진다는 장점과 연속적인 집계연산자(aggregate operator)를 사용할 수 있다는 장점이 있다. 또한 앞에서 지적한 이중 언어 정의 문제를 해결할 수 있다.

• Journal of KIISE:Databases
• /
• v.36 no.4
• /
• pp.280-291
• /
• 2009

Skyline Query processing is important to wireless sensor applications in order to process multi-dimensional data efficiently. Most skyline researches about sensor network focus on minimizing the energy consumption due to the battery powered constraints. In order to reduce energy consumption, Filtering Method is proposed. Most existing researches have assumed a snapshot skyline query processing and do not consider continuous queries and use data generated in ancestor node. In this paper, we propose an energy efficient method called Bottom up filtering tuple selection for continuous skyline query processing. Past skyline data generated in child nodes are stored in each sensor node and is used when choosing filtering tuple. We also extend the algorithms, called Support filtering tuple(SFT) that is used when we choose the additional filtering tuple. There is a temporal correlation between previous sensing data and recent sensing data. Thus, Based on past data, we estimate current data. By considering this point, we reduce the unnecessary communication cost. The experimental results show that our method outperforms the existing methods in terms of both data reduction rate(DRR) and total communication cost.