Vol. 47, No. 7,
Jul. 2020
Digital Library
Winner Determination of Combinatorial Auction Using the Knapsack Problem
http://doi.org/10.5626/JOK.2020.47.7.629
In combinatorial auctions, bidders can make bids on a set of items. When the items show complementarities in values, combinatorial auctions make it possible to express the values more accurately and enhance the market’s efficiency. The winner determination problem is an NP-complete problem. To address this difficult and important problem, we can attempt to solve the special subproblem that can be solved efficiently or develop approximate solutions. In this paper, we reformulated the WDP into a knapsack problem. By using Markov chain Monte Carlo method, we achieved a desired stationary solution with a large objective function. The solution is comparable to that of integer programming.
PSL-DB: Non-Volatile Memory-optimized LSM-Tree with Skip List
Chanyeol Park, Dongui Kim, Beomseok Nam
http://doi.org/10.5626/JOK.2020.47.7.635
With the release of Intel"s Optane DC Persistent Memory, non-volatile memory, offering higher capacity than DRAM and showing higher performance than SSD and HDD, is in the spotlight as the next generation of storage devices. In this paper, we propose the Persistent Skip List DataBase (PSL-DB), a key-value store system optimized for the Optane DCPM in app-direct mode. PSL-DB uses a byte-addressable skip list that significantly reduces the I/O traffic as it avoids redundant writes. PSL-DB also does not sacrifice write performance for read performance as it does not degrade the write performance via artificial governors. In our experiments using Intel Optane DC Persistent Memory, PSL-DB shows significantly higher query processing throughput than legacy LevelDB that stores SSTables in Optane DC PM.
Adjusting OS Scheduler Parameters to Improve Server Application Performance
Taehyun Han, Hyeonmyeong Lee, Heeseung Jo
http://doi.org/10.5626/JOK.2020.47.7.643
Modern Linux servers are used in a variety of ways, from large servers to small IOTs, and most machines run their services through the default scheduler provided by Linux. Although it is possible to optimize for a specific purpose, there is a problem in which the average user cannot optimize all modern Linux applications. In this paper, we propose SCHEDTUNE to automatically optimize the scheduler configuration to maximize Linux server performance. SCHEDTUNE allows users to improve performance without modification to the application or basic kernel source running on the server. This makes it easy for administrators to configure schedulers that operate specifically for their servers. Experimental results showed that when SCHEDTUNE is applied, the maximum performance is achieved up to 19 %, and in most cases performance improvement is achieved as well.
Usability Assessment of FHIR-based Geriatric Depression Scale Questionnaire Using Chatbot
http://doi.org/10.5626/JOK.2020.47.7.650
As Korea enters the aging society, the interest in, and importance of the elderly are increasing. In particular, the depression of the elderly is an important issue to be addressed. To this end, latency delays are among the most common complaints about those who seek medical examination or to see a doctor. Also, if patients move and are thus are sent to a different hospital because of a change of residence or personal reasons, they may undergo the same examination. In this case, fatigue and economic burden are placed upon the patient because of the re-examination. In this study, we have implemented the chatbot for the Geriatric Depression Scale Questionnaire based on the Fast Healthcare Interoperability Resource, an international health information exchange standard. Unlike the existing paper questionnaire, it has interoperable questionnaire information, and the user’s perceived usability was examined through the evaluation of usability.
The Cut Transition Detection Model Using the SSD Method
Sungmin Park, Ui Nyoung Yoon, Geun-Sik Jo
http://doi.org/10.5626/JOK.2020.47.7.655
Shot boundary detection is constantly being studied as an essential technique for analyzing video content. In this paper, we propose an End-to-End Learning model using the SSD (Single Shot Multibox Detector) method to resolve the shortcomings of the existing research and to identify the exact location of the cut transition. We applied the concept of the Multi-Scale Feature Map and Default box of the SSD to predict multiple cut transitions, and combined the concept of Image Concatenation, one of the image comparison methods, with the model to reinforce the feature information of the cut transitions. The proposed model showed 88.7% and 98.0% accuracy in the re-labeled ClipShots and TRECVID 2007 datasets, respectively, compared to the latest research. Additionally, it detected a range closer to the correct answer than the existing deep learning model.
CNN-based Speech Emotion Recognition Model Applying Transfer Learning and Attention Mechanism
Jung Hyun Lee, Ui Nyoung Yoon, Geun-Sik Jo
http://doi.org/10.5626/JOK.2020.47.7.665
Existing speech-based emotion recognition studies can be classified into the case of using a voice feature value and a variety of voice feature values. In the case of using a voice feature value, there is a problem that it is difficult to reflect the complex factors of the voice such as loudness, overtone structure, and range of voices. In the case of using various voice feature values, studies based on machine learning comprise a large number, and there is a disadvantage in that emotion recognition accuracy is relatively lower than that of deep learning-based studies. To resolve this problem, we propose a speech emotion recognition model based on a CNN(Convolutional Neural Network) using Mel-Spectrogram and Mel Frequency Cepstral Coefficient (MFCC) as voice feature values. The proposed model applied transfer learning and attention to improve learning speed and accuracy, and achieved 77.65% emotion recognition accuracy, showing higher performance than the comparison works.
OWL-Horst Ontology Inference Engine Using Distributed Table Structure in Cloud Computing Environment
Min-Sung Kim, Min-Ho Lee, Wan-Gon Lee, Young-Tack Park
http://doi.org/10.5626/JOK.2020.47.7.674
Recently, many machine learning methods that extend ontology through data obtained from the web are being studied. As data from the web continues to increase, interest in large-capacity ontology inference methods is also increasing. However, the increasing amount of data decreases processing speeds. This paper describes how to improve the performance of large-scale OWL-Horst inference using distributed table structured data frames to solve the problem of the slow processing speed of large-capacity data. Also, a distributed parallel inference algorithm and optimization method used to improve the inference performance is described. To evaluate the performance of the inference system using the distributed table structured data frame proposed in this paper, experiments were conducted with LUBM1000, LUBM2000, LUBM3000, and LUBM4000. Our reasoning system showed the best performance.
A Small-Scale Korean-Specific BERT Language Model
Sangah Lee, Hansol Jang, Yunmee Baik, Suzi Park, Hyopil Shin
http://doi.org/10.5626/JOK.2020.47.7.682
Recent models for the sentence embedding use huge corpus and parameters. They have massive data and large hardware and it incurs extensive time to pre-train. This tendency raises the need for a model with comparable performance while economically using training data. In this study, we proposed a Korean-specific model KR-BERT, using sub-character level to character-level Korean dictionaries and BidirectionalWordPiece Tokenizer. As a result, our KR-BERT model performs comparably and even better than other existing pre-trained models using one-tenth the size of training data from the existing models. It demonstrates that in a morphologically complex and resourceless language, using sub-character level and BidirectionalWordPiece Tokenizer captures language-specific linguistic phenomena that the Multilingual BERT model missed.
A Technique of Protecting User Sensitive Partial Trajectory with Local Differential Privacy on the Road Network
http://doi.org/10.5626/JOK.2020.47.7.693
Today, with the proliferation of smartphones and the development of sensor technology, path data, a list of user location data collected from mobile devices, is being manipulated for marketing or efficient algorithm development. However, such indiscriminate collection of location information may cause personal privacy leakage issues. To resolve the problem, many differential privacy techniques have been proposed. However, the previous methods significantly degrade query accuracy if they are applied in the trajectory dataset. Additionally, the differential privacy technique is classified into a curator model and a local model. The local model has advantages of not having a reliable server, but suffers from more noise inserted to reduce query accuracy. This paper classifies vertices into heavy points and light points to resolve the problem of data usability in applying differential privacy to collect road network trajectory data in the local model. Additionally, experiments show that the proposed technique mitigates the degradation of overall data usability while protecting the sensitive data in accordance with the differential privacy standards.
Deploying UAV based on Reinforcement Learning for Throughput Maximization in UAV Environments
http://doi.org/10.5626/JOK.2020.47.7.700
Because of the commercialization of the 5G network, many base stations must enhance a reliable communication quality. Thus, many studies are being conducted to provide mobility and economic benefits to the UAVs-Base Station (UAVs-BS) on behalf of the ground base stations. In this paper, we propose a system to identify a location wherein multiple users can access optimal service throughput by considering users’ requirements and the Base Station(BS)’s position in UAVs communication. Based on the Air-To-Ground(A2G) Path Loss Model, the virtual communication environment is established and Max-Min Airtime Fairness is applied for equitable channel usage time distribution according to user requirements. Additionally, the Proximal Policy Optimization (PPO) algorithm is applied to set an optimal location with the maximum throughput. As a result, the proposed systems allow the UAVs to be in the locations with high service throughput for users with different demands.
Search
Journal of KIISE
- ISSN : 2383-630X(Print)
- ISSN : 2383-6296(Electronic)
- KCI Accredited Journal
Editorial Office
- Tel. +82-2-588-9240
- Fax. +82-2-521-1352
- E-mail. chwoo@kiise.or.kr