• Title/Summary/Keyword: Threshold setting

Search Result 150, Processing Time 0.023 seconds

A Study on the Improvement of Prediction Accuracy of Collaborative Recommender System under the Effect of Similarity Weight Threshold (협력적 추천시스템에서 유사도 가중치의 임계치 설정에 따른 선호도 예측 정확도 향상에 관한 연구)

  • Lee, Seok-Jun
    • Korean Business Review
    • /
    • v.20 no.1
    • /
    • pp.145-168
    • /
    • 2007
  • Recommender system helps customers to find easily items and helps the e-biz companies to set easily their target customer by automated recommending process. Recommender systems are being adopted by several e-biz companies and from these systems, both of customers and companies take some benefits. This study sets several thresholds to the similarity weight, which indicates a degree of similarity of two customers' preference, to improve the performance of prediction accuracy. According to the threshold, the accuracy of prediction is being improved but some threshold setting shows the reduction of the prediction rate, which is the coverage. This coverage reduction has male effect on the prediction accuracy of customers, so more study on the prediction accuracy of recommender system and to maximize the coverage are needed.

  • PDF

Index of union and other accuracy measures (Index of Union와 다른 정확도 측도들)

  • Hong, Chong Sun;Choi, So Yeon;Lim, Dong Hui
    • The Korean Journal of Applied Statistics
    • /
    • v.33 no.4
    • /
    • pp.395-407
    • /
    • 2020
  • Most classification accuracy measures for optimal threshold are divided into two types: one is expressed with cumulative distribution functions and probability density functions, the other is based on ROC curve and AUC. Unal (2017) proposed the index of union (IU) as an accuracy measure that considers two types to get them. In this study, ten kinds of accuracy measures (including IU) are divided into six categories, and the advantages of the IU are studied by comparing the measures belonging to each category. The optimal thresholds of these measures are obtained by setting various normal mixture distributions; subsequently, the first and second type of errors as well as the error sums corresponding to each threshold are calculated. The properties and characteristics of the IU statistic are explored by comparing the discriminative power of other accuracy measures based on error values.The values of the first type error and error sum of IU statistic converge to those of the best accuracy measures of the second category as the mean difference between the two distributions increases. Therefore, IU could be an accuracy measure to evaluate the discriminant power of a model.

Study on Acquisition Method for GAS-CP-CDMA (Cyclic-Prefix Code Division Multiple Access using Good-Autocorrelation-Sequence) (GAS-CP-CDMA 동기 포착 방법 연구)

  • Lee, Seong-Min;Kim, Young-Je;Kim, Whan-Woo
    • The Journal of Korean Institute of Communications and Information Sciences
    • /
    • v.33 no.11A
    • /
    • pp.1087-1094
    • /
    • 2008
  • In this paper, we propose the acquisition method of the Cyclic-Prefix CDMA using Good-Autocorrelation-Sequence (GAS-CP-CDMA) system. The sequences for GAS-CP-CDMA are m-sequence and constant amplitude zero autocorrelation code (CAZAC) with good autocorrelation property. The GAS-CP-CDMA system is proposed for interference cancellation in multi-user and -path environment in co-cell. However, the reliable initial acquisition in multi-path fading channel and multi-user system environment is crucial. In this paper, non-coherent power detector and threshold value are used for discriminating signal acquisition. The sequence used for computer simulation is CAZAC. The simulation results show proper threshold value depends on the number of users. Through the computer simulation, we draw setting up method of the optimum threshold value for the GAS-CP-CDMA system.

Accuracy Evaluation of UHF Wind Profiler Radar Wind Vectors by Setting a Threshold of Signal-to-Noise Ratios (신호대잡음비의 임계값 설정에 따른 UHF 윈드프로파일러 바람벡터의 정확도 평가)

  • Kim, Kwang-Ho;Kim, Park-Sa;Kim, Min-Seong;Kang, Dong-Hwan;Kwon, Byung Hyuk
    • Journal of Environmental Science International
    • /
    • v.25 no.9
    • /
    • pp.1241-1251
    • /
    • 2016
  • A minimum threshold for the signal to noise ratio ($SNR_{min}$) has to be set in the data processing system of wind profiler radar (WPR). The data collection rate and the accuracy of the WPR wind vector depend on the $SNR_{min}$. The WPR at Uljin is operated with an $SNR_{min}$ of 1 dB which is a relatively large threshold. We found that the accuracy and the continuity of the WPR wind vector with height were directly related to the variability of the SNR and vertical gradient of the squared refractive index. We investigated a quantitative method for determining a new $SNR_{min}$ for the WPR at Uljin and it was evaluated with radiosonde data. The accuracy and continuity of the wind vector from an SNR of less than 1 dB, began to decrease at an altitude of 3.5 km. Most of the SNR values were less than -3.5 dB in altitudes higher than 3.5 km. We retrieved high-accuracy wind vectors at altitudes over 3 km where measurements were deficient with an $SNR_{min}$ of 1 dB.

Additive Noise Reduction Algorithm for Mass Spectrum Analyzer (질량 스펙트럼 분석기를 위한 부가잡음제거 알고리즘)

  • Choi, Hun;Lee, Imgeun
    • Journal of the Korea Institute of Information and Communication Engineering
    • /
    • v.22 no.1
    • /
    • pp.33-39
    • /
    • 2018
  • An additive noise reduction algorithm for a mass spectrum analyzer is proposed. From the measured ion signal, we first used an estimated threshold from the mode of the measured signal to eliminate background noises with the white Gaussian characteristics. Also, a signal block corresponding to each mass index is constructed to perform a second order curve fitting and a linear approximation to signal block. In this process, the effective signal block composed of only the ion signal can be reconstructed by removing the impulsive noises and the sample signals which are insufficient to be viewed as normal ion signals. By performing curve fitting on the effective signal block, the noise-free mass spectrum can be obtained. To evaluate the performance of the proposed method, a simulation was performed using the signals acquired from the development equipment. Simulation results show the validity of the threshold setting from the mode and the superiority of the proposed curve fitting and linear approximation based noise canceling algorithm.

Data Transmission System Applying An Adaptive Threshold Based Multi-channel Sound (적응적 임계치를 적용한 멀티 채널 소리 기반의 데이터 전송 시스템)

  • Gang, Hyun-Mo;Jung, Jin-Woo;Choi, Chun-Yong;Kwon, Young-Hun;Lee, Sung-Koo
    • Journal of Digital Contents Society
    • /
    • v.15 no.1
    • /
    • pp.93-99
    • /
    • 2014
  • Recently Wireless communication among short-distance devices has come to notice due to smart phone generalization recently. However, instead of setting up additional H/W, communication technology providing wireless communication based on S/W is in need due to limited availability of NFC's use. Accordingly, short-distance wireless communication technology that makes great use of mike and speaker which installed in every device draws attention. Our thesis suggests improvement of acoustic transmission speed by applying multi-channel parallel transmission and advancement of transmission rate that differed from each mike's own characteristics through optimizing adaptive threshold. The study is not only just applied in specific and limited conditions such as promoting corporation and payments system but also fast and convenient data transmit system general users-oriented.

Optimal Adaptive Multiband Spectrum Sensing in Cognitive Radio Networks

  • Yu, Long;Wu, Qihui;Wang, Jinlong
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • v.8 no.3
    • /
    • pp.984-996
    • /
    • 2014
  • In this paper, optimal sensing time allocation for adaptive multiband spectrum sensing-transmission procedure is investigated. The sensing procedure consists of an exploration phase and a detection phase. We first formulate an optimization problem to maximize the throughput by designing not only the overall sensing time, but also the sensing time for every stage in the exploration and detection phases, while keeping the miss detection probability for each channel under a pre-defined threshold. Then, we transform the initial non-convex optimization problem into a convex bilevel optimization problem to make it mathematically tractable. Simulation results show that the optimized sensing time setting in this paper can provide a significant performance gain over the previous studies.

Transcranial Doppler Detection of Vasospasm Following Subarachnoid Hemorrhage (지주막하 출혈에 따른 Vasospasm에 대한 Transcranial Doppler의 임상적 적용)

  • Lee, Jun Hong
    • Annals of Clinical Neurophysiology
    • /
    • v.1 no.1
    • /
    • pp.55-59
    • /
    • 1999
  • Vasoconstriction of intracerebral arteries is the leading cause of delayed cerebral infarction and mortality following aneurysmal subarachnoid hemorrhage. Transcranial Doppler studies show and increase in the flow velocities of basal cerebral arteries, which usually start around day 4 following a subarachnoid hemorrhage, and peaking by days 7 to 14. Angiographic studies confirm the presence of at least some degree of MCA vasospasm when the flow velocities are higher than 100 cm/sec. Mean velocities in the 120 to 200 cm/sec range correspond to 25 to 50% luminal narrowing. MCA and ACA vsospasm is detected with around 90% specificity. Sensitivity is 80% and 50% respectively. A 200cm/sec threshold and rapid flow velocity increase exceeding 50 cm/sec on consecutive days, has been associated with subsequent infarction. Transcranial Doppler is also used to monitor the effects of endovascular treatment of vasospasm. Flow velocities decrease following successful angioplasty or papaverine infusion. Overall, transcranial Doppler studies are considered to have acceptable accuracy for the evaluation of vasospasm in aneurysmal subarachnoid hemorrhage, with limitations that have to be taken into consideration in the clinical setting.

  • PDF

Distance Measurement by Automatic Peak Detection for Indoor Positioning Using Spread Spectrum Ultrasonic Waves

  • Suzuki, Akimasa;Miyara, Yasuaki;Iyota, Taketoshi;Kim, Young-Bok;Choi, Yong-Woon
    • Journal of Power System Engineering
    • /
    • v.19 no.2
    • /
    • pp.33-39
    • /
    • 2015
  • In conducting indoor positioning by code division multiple access using spread spectrum ultrasonic waves, it is required to detect signals under the influence of near-far problem occurred by difference on signal power, caused by distance between transmitter and receiver. For discussing robustness to the problem, we verified measuring accuracy on distance from an experiment on a real space with a hardware device where our proposed method is mounted. The proposed method performs automatic signal detection by setting threshold level dynamically. As an experimental result, measurable distance were improved by the proposed method, and measurement errors were up to 50mm in distances from 1000mm to 6000mm; therefore, enough accuracy to realize self-localization or navigation for autonomous mobile robot or human was obtained.

A Study on locomotion of a mobile robot by a visual perception (시각정보에 의한 이동 로보트의 주행에 관한 연구)

  • Shin, J.S.;Jeong, D.M.;Cho, J.M.;Chang, W.S.;Hong, S.H.
    • Proceedings of the KIEE Conference
    • /
    • 1987.07b
    • /
    • pp.1260-1263
    • /
    • 1987
  • This paper describes the mobile robot system to recognize the guidance tape, and presents the locomotion algorithm. This system converts video imago to binary image by setting an optimal threshold and obtains the parameters to move the robot. The mobile robot moves according to the programmed route in memory. But after recognized the obstacle on the locomotion routs, this system constructs the new route and the robot moves following the new route.

  • PDF