• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.1
• /
• pp.91-98
• /
• 2023

With the increasing flood damage by frequently localized heavy rains, flood prediction research are being conducted to prevent flooding damage in advance. In this paper, we present a machine-learning scheme for developing a flooding depth and location prediction model using real-time rainfall data. This scheme proposes a dataset configuration method using the data as input, which can robustly configure various rainfall distribution patterns and train the model with less memory. These data are composed of two: valid total data and valid local. The one data that has a significant effect on flooding predicted the flooding location well but tended to have different values for predicting specific rainfall patterns. The other data that means the flood area partially affects flooding refers to valid local data. The valid local data was well learned for the fixed point method, but the flooding location was not accurately indicated for the arbitrary point method. Through this study, it is expected that a lot of damage can be prevented by predicting the depth and location of flooding in a real-time manner.

• The KIPS Transactions:PartD
• /
• v.8D no.6
• /
• pp.813-822
• /
• 2001

In this paper we proposed four data migration methods based on time segmented storage structure including past segment, current segment, and future segment. The migration methods proposed in this paper are the Time Granularity migration method, the LST-GET (Least valid Start Time-Greatest valid End Time) migration method, the AST-AET (Average valid Start Time-Average valid End Time) migration method, and the Min-Overlap migration method. In the each data migration method we define the dividing criterion among segments and entity versions to store on each segment. We measured the response time of queries for the proposed migration methods. When there are no LLTs (Long Lived Tuples), the average response time of AST-AET migration method and LST-GET migration method are smaller than that of Time Granularity migration method. In case of existing LLT, the performance of the LST-GET migration method decreased. The AST-AET migration method resulted in better performance for queries than the Time Granularity migration method and the LST-GET migration method. The Min-Overlap migration method resulted in the almost equal performance for queries compared with the AST-AET migration method, in case of storage utilization more efficient than the AST-AET.

• Journal of KIISE:Databases
• /
• v.28 no.3
• /
• pp.384-394
• /
• 2001

In this paper, we propose AST-AET(Average valid Start Time-Average valid End Time) data migration strategy based on the storage structure where temporal data is divided into a past segment, a current segment, and a future segment. We define AST and AET which are used in AST-AET data migration strategy and also define entity versions to be stored in each segment. We describe methods to compute AST and AET, and processes to search entity versions for migration and move them. We compare average response times for user queries between AST-AET data migration strategy and the existing LST-GET(Least valid Start Time-Greatest valid End Time) data migration strategy. The experimental results show that, when there are no LLTs(Long Lived Tuples), there is little difference in performance between the two migration strategies because the size of a current segment is nearly equal. However, when there are LLTs, the average response time of AST-AET data migration strategy is smaller than that of LST-GET data migration strategy because the size of a current segment of LST-GET data migration strategy becomes larger. In addition, when we change average interarrival times of temporal queries, generally the average response time of AST-AET data migration strategy is smaller than that of LST-GET data migration strategy.

• Journal of Internet of Things and Convergence
• /
• v.8 no.1
• /
• pp.37-43
• /
• 2022

The fundamental basis of AI technology is learningable data. Recently, the types and amounts of data collected and produced by the government or private companies are increasing exponentially, however, verified data that can be used for actual machine learning has not yet led to it. This study discusses the conditions that data actually can be used for machine learning should meet, and identifies factors that degrade data quality through case studies. To this end, two representative cases of developing a prediction model using public big data was selected, and data for actual problem solving was collected from the public data portal. Through this, there is a difference from the results of applying valid data screening criteria and post-processing. The ultimate purpose of this study is to argue the importance of data quality management that must be most fundamentally preceded before the development of machine learning technology, which is the core of artificial intelligence, and accumulating valid data.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.12
• /
• pp.2947-2959
• /
• 1997

Temporal data model is able to handle the time varying information, which is to add temporal attributes to conventional data model. The temporal data model is classified into three models depending upon supporting time dimension, that are the valid time model to support valid time, the transaction time model to support transaction model, and the bitemporal data model to support valid time and transaction time. Most temporal data models are designed to process the temporal data by extending the relational model. There are two types or temporal data model, which are the tuple timestamping and the attribute timestamping depending on time dimension. In this research, a concepts of temporal data model, the time dimension, types of thc data model, and a consideration for the data model design are discussed Also, temporal data models in terms of the time dimension are compared. And the aggregate function model of valid time model is proposed, and then logical analysis for its computing consts has been done.

• International Journal of Precision Engineering and Manufacturing
• /
• v.9 no.4
• /
• pp.79-82
• /
• 2008

A method is described for ideally reconstructing the profile from a surface profiling measurement containing a reasonable amount of null measurement data. The proposed method can conjecture lost information and rectify irregular data that result due to bad measuring environments, signal transmission noise, or instrument-induced errors, The method adopts the concept of computer graphics and consists of several processing steps. First, a search for valid data in the neighborhood of the null data is performed. The valid data are then grouped and their contours are extracted. By analyzing these contours, a bounding box can be obtained and the general distribution of the entire area encompassing the valid and null data is determined Finally, an ideal surface model is overlaid onto the measurement results based on the bounding box, generating a complete reconstruction of the calculations, A surface-profiling task on a liquid crystal display photo spacer is used to verify the proposed method. The results are compared to those obtained through the use of a scanning electron microscope to demonstrate the accuracy of the proposed method.

• Science of Emotion and Sensibility
• /
• v.20 no.3
• /
• pp.39-48
• /
• 2017

In order to verify the effectiveness of marketing in the basketball stadium, this study measured and analyzed the gaze frequency and interest when the pupil was expanded by using the eye-tracking technology among various neuro marketing techniques of marketing. To analyze the section where the pupil size get expanded, interval of pupil size was higher than 2.275% (2 sigma data) and higher than 0.135% high (3 sigma data). Overall the valid data was analyzed by inflection points according to gaze frequency. We also analyzed the correlation between overall valid data and the ranges where the pupil size was significantly increased. The result showed that the correlation between overall valid data and pupil size 2 sigma data showed the highest correlation with 0.805. The pupil size 2 sigma data and pupil size 3 sigma data showed a correlation with 0.781, overall the valid data and pupil size 2 sigma data showed a correlation with 0.683. Therefore, it is concluded that, the section where the pupil size was expanded and the section at which gaze frequency is higher in the eye-tracking data were similar. However, the correlation between data of pupil size is determined to be significantly expanded and overall the valid data is decreased.

• Wind and Structures
• /
• v.19 no.3
• /
• pp.271-294
• /
• 2014

The impact of artefacts in archived wind observations on the design wind speed obtained by extreme value analysis is demonstrated using case studies. A signpost protocol for detecting candidate artefacts is described and its performance assessed by comparing results against previously validated data. The protocol targets artefacts by exploiting the serial correlation between observations. Additional "sieve" algorithms are proposed to identify types of correctable artefact from their "signature" in the data. In extreme value analysis, artefacts displace valid observations only when they are larger, hence always increase the design wind speed. Care must be taken not identify large valid values as artefacts, since their removal will tend to underestimate the design wind speed.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.42 no.3
• /
• pp.25-34
• /
• 2017

K-means is a popular and efficient data clustering method that only uses intra-cluster distance to establish a valid index with a previously fixed number of clusters. K-means is useless without a suitable number of clusters for unsupervised data. This paper aimsto propose the Group Search Optimization (GSO) using Silhouette to find the optimal data clustering solution with a number of clusters for unsupervised data. Silhouette can be used as valid index to decide the number of clusters and optimal solution by simultaneously considering intra- and inter-cluster distances. The performance of GSO using Silhouette is validated through several experiment and analysis of data sets.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.5 no.4
• /
• pp.243-253
• /
• 2010

Multihop data delivery in vehicular ad hoc networks (VANETs) suffers from the fact that vehicles are highly mobile and inter-vehicle links are frequently disconnected. In such networks, for efficient multihop routing of road safety information (e.g. road accident and emergency message) to the area of interest, reliable communication and fast delivery with minimum delay are mandatory. In this paper, we propose a multihop vehicle-to-infrastructure routing protocol named Vertex-Based Predictive Greedy Routing (VPGR), which predicts a sequence of valid vertices (or junctions) from a source vehicle to fixed infrastructure (or a roadside unit) in the area of interest and, then, forwards data to the fixed infrastructure through the sequence of vertices in urban environments. The well known predictive directional greedy routing mechanism is used for data forwarding phase in VPGR. The proposed VPGR leverages the geographic position, velocity, direction and acceleration of vehicles for both the calculation of a sequence of valid vertices and the predictive directional greedy routing. Simulation results show significant performance improvement compared to conventional routing protocols in terms of packet delivery ratio, end-to-end delay and routing overhead.