Audio Data Mining Using Multi-perceptron Artificial Neural Network
|
|
|
- Naomi O’Connor’
- 10 years ago
- Views:
Transcription
1 224 IJCSNS Internatonal Journal of Computer Scence and Network Securty, VOL.8 No.0, October 2008 Audo Data Mnng Usng Mult-perceptron Artfcal Neural Network Surendra Shetty, 2 K.K. Achary Dept of Computer Applcatons, NMAM Insttute Of Technology, Ntte,Udup, Karnataka, Inda, PIN- 5740, 2 Dept of Statstcs, Mangalore Unversty, Mangalagangothr, Inda, PIN-57499, Abstract--Data mnng s the actvty of analyzng a gven set of data. It s the process of fndng patterns from large relatonal databases. Data mnng ncludes: extract, transform, and load transacton data onto the data warehouse system, store and manage the data n a multdmensonal database system, provdes data, analyze the data by applcaton software and vsual presentaton. Audo data contans nformaton of each audo fle such as sgnal processng component- power spectrum, cepstral values that s representatve of partcular audo fle. The relatonshp among patterns provdes nformaton. It can be converted nto knowledge about hstorcal patterns and future trends. Ths work nvolves n mplementng an artfcal neural network (ANN) approach for audo data mnng. Acqured audo s preprocessed to remove nose followed by feature extracton usng cepstral method. The ANN s traned wth the cepstral values to produce a set of fnal weghts. Durng testng process (audo mnng), these weghts are used to mne the audo fle. In ths work, 50 audo fles have been used as an ntal attempt to tran the ANN. The ANN s able to produce only about 90% accuracy of mnng due to less correlaton of audo data. Index Terms ANN, Backpropagaton Algorthm, Cepstrum, Feature Extracton, FFT, LPC, Perceptron, Testng, Tranng, Weghts I. INTRODUCTION Data mnng s concerned wth dscoverng patterns meanngfully from data. Data mnng has deep roots n the felds of statstcs, artfcal ntellgence, and machne learnng. Wth the advent of nexpensve storage space and faster processng over the past decade, the research has started to penetrate new grounds n areas of speech and audo processng as well as spoken language dalog. It has ganed nterest due to audo data that are avalable n plenty. Algorthmc advances n automatc speech recognton have also been a major, enablng technology behnd the growth n data mnng. Currently, large vocabulary, contnuous speech recognzers are now traned on a record amount of data such as several hundreds of mllons of words and thousands of hours of speech. Poneerng research n robust speech processng, large-scale dscrmnatve tranng, fnte state automata, and statstcal hdden Markov modelng have resulted n real-tme recognzers that are able to transcrbe spontaneous speech. The technology s now hghly attractve for a varety of speech mnng applcatons. Audo mnng research ncludes many ways of applyng machne learnng, speech processng, and language processng algorthms []. It helps n the areas of predcton, search, explanaton, learnng, and language understandng. These basc challenges are becomng ncreasngly mportant n revolutonzng busness processes by provdng essental sales and marketng nformaton about servces, customers, and product offerngs. A new class of learnng systems can be created that can nfer knowledge and trends automatcally from data, analyze and report applcaton performance, and adapt and mprove over tme wth mnmal or zero human nvolvement. Effectve technques for mnng speech, audo, and dalog data can mpact numerous busness and government applcatons. The technology for montorng conversatonal audo to dscover patterns, capture useful trends, and generate alarms s essental for ntellgence and law enforcement organzatons as well as for enhancng call center operaton. It s useful for a dgtal object dentfer analyzng, montorng, and trackng customer preferences and nteractons to better establsh customzed sales and techncal support strateges. It s also an essental tool n meda content management for searchng through large volumes of audo warehouses to fnd nformaton, documents, and news. II. TECHNICAL WORK PREPARATION A. Problem statement Audo fles are to be mned properly wth hgh accuracy gven partal audo nformaton. Ths can be very much acheved usng ANN. Ths work nvolves n mplementng supervsed backpropagaton algorthm(bpa). The BPA s traned wth the features of audo data for dfferent number of nodes n the hdden layer. The layer wth optmal number of nodes has to be chosen for proper audo mnng. Manuscrpt receved October 5, 2008 Manuscrpt revsed October 20, 2008
2 IJCSNS Internatonal Journal of Computer Scence and Network Securty, VOL.8 No.0, October B. Overvew of Audo mnng Audo recognton s a classc example of thngs that the human bran does well, but dgtal computers do poorly. Dgtal computers can store and recall vast amounts of data perform mathematcal calculaton at blazng speeds and do repettve tasks wthout becomng bored or neffcent. Computer performs very poorly when faced wth raw sensory data. Teachng the same computer to understand audo s a major undertakng. Dgtal sgnal processng generally approaches the problem of audo recognton nto two steps, ) Feature extracton, 2) Feature matchng Each word n the ncomng audo sgnal s solated and then analyzed to dentfy the type of exctaton and resonate frequency [2]. These parameters are then compared wth prevous example of spoken words to dentfy the closest match. Often, these systems are lmted to few hundred words; can only accept sgnals wth dstnct pauses between words; and must be retraned. Whle ths s adequate for many commercal applcatons, these lmtatons are humblng when compared to the abltes of human hearng. There are two man approaches to audo mnng.. Text-based ndexng: Text-based ndexng, also known as large-vocabulary contnuous speech recognton, converts speech to text and then dentfes words n a dctonary that can contan several hundred thousand entres. 2. Phoneme-based ndexng: Phoneme based ndexng doesn t convert speech to text but nstead works only wth sounds. The system frst analyzes and dentfes sounds n a pece of audo content to create a phonetc-based ndex. It then uses a dctonary of several dozen phonemes to convert a user s search term to the correct phoneme strng. (Phonemes are the smallest unt of speech n a language, such as the long a sound that dstngushes one utterance from another. All words are sets of phonemes). Fnally, the system looks for the search terms n the ndex. A phonetc system requres a more propretary search tool because t must phonetcze the query term, and then try to match t wth the exstng phonetc strng output. Although audo mnng developers have overcome numerous challenges, several mportant hurdles reman. Precson s mprovng but t s stll a key ssue mpedng the technology s wdespread adopton, partcularly n such accuracy-crtcal applcatons as court reportng and medcal dctaton. Audo mnng error rates vary wdely dependng on factors such as background nose and cross talk. Processng conversatonal speech can be partcularly dffcult because of such factors as overlappng words and background nose [3][4]. Breakthroughs n natural language understandng wll eventually lead to bg mprovements. The problem of audo mnng s an area wth many dfferent applcatons. Audo dentfcaton technques nclude, Channel vocoder Lnear predcton Formant vocodng Cepstral analyss There are many current and future applcatons for audo mnng. Examples nclude telephone speech recognton systems, or voce dalers on car phones. C. Schematc dagram The sequence of Audo mnng can be schematcally shown as below... Sequence of audo processng D. Artfcal neural network A neural network s constructed by hghly nterconnected processng unts (nodes or neurons) whch perform smple mathematcal operatons [5]. Neural networks are characterzed by ther topologes, weght vectors and actvaton functon whch are used n the hdden layers and output layer [6]. The topology refers to the number of hdden layers and connecton between nodes n the hdden layers. The actvaton functons that can be used are sgmod, hyperbolc tangent and sne [7]. A very good account of neural networks can be found n []. The network models can be statc or dynamc [8]. Statc networks nclude sngle layer perceptrons and multlayer perceptrons. A perceptron or adaptve lnear element (ADALINE) [9] refers to a computng unt. Ths forms the basc buldng block for neural networks. The nput to a perceptron s the summaton of nput pattern vectors by weght vectors. In Fgure 2, the basc functon of a sngle layer perceptron s shown. Fg
3 226 IJCSNS Internatonal Journal of Computer Scence and Network Securty, VOL.8 No.0, October 2008 Fg. 2. Operat on of a neuron In Fgur e 3, a mult layer perceptron s shown schematcally. Informaton flows n a feed-forward manner from nput layer to the output layer through hdden layers. The number of nodes n the nput layer and output layer s fxed. It depends upon the number of nput varables and the number of output varables n a pattern. In ths work, there are sx nput varables and one output varable. The number of nodes n a hdden layer and the number of hdden layers are varable. Dependng upon the type of applcaton, the network parameters such as the number of nodes n the hdden layers and the number of hdden layers are found by tral and error method [0] g. 3 Multl ayer percep tron In most of the appl cato ns one hdde n layer s suffcent. The actvaton functon whch s used to tran the ANN, s the sgmod functon and t s gven by: f(x) = + exp( -x) () where, f (x) s a non - lnear dfferentable functon, Nn x = å W = j (p) x n (p) + θ (p), where, N n s the total number of nodes n the n th layer F W j s the weght vector connectng th neuron of a layer wth the j th neuron n the next layer. q s the threshold appled to the nodes n the hdden layers and output layer and p s the pattern number. In the frst hdden layer, x s treated as an nput pattern vector and for the successve layers, x s the output of the th neuron of the proceedng layer. The output x of a neuron n the hdden layers and n the output layer s calculated by: X n + (p) = (2) + exp( - x + θ(p)) For each pattern, error E (p) n the output layers s calculated by: E(p) = 2 N M M 2 Σ (d (p) - x (p)) (3) = where M s the total number of layer whch nclude the nput layer and the output layer, N M s the number of nodes n the output layer. d (p) s the desred output of a pattern and X M (p) s the calculated output of the network for the same pattern at the output layer. The total error E for all patterns s calculated by: E = å L (4) E(p) p= where, L s the total number of patterns. E. Implementaton The flowchart, fg. 4 explans the sequence of mplementaton of audo mnng. Ffty audo fles were chosen. The feature extracton procedure s appled Preemphaszng and wndowng. Audo s ntrnscally a hghly non-statonary sgnal. Sgnal analyss, FFT-based or Lnear Predctor Coeffcents (LPC) based, must be carred out on short segments across whch the audo sgnal s assumed to be statonary. The feature extracton s performed on 20 to 30 ms wndows wth 0 to 5 ms shft between two consecutve wndows. To avod problems due to the truncaton of the sgnal, a weghtng wndow wth the approprate spectral propertes must be appled to the analyzed chunk of sgnal. Some wndows are Hammng, Hannng and Blackman.
4 IJCSNS Internatonal Journal of Computer Scence and Network Securty, VOL.8 No.0, October Normalzaton Feature normalzaton can be used to reduce the msmatch between sgnals recorded n dfferent condtons. Normalzaton conssts n mean removal and eventually varance normalzaton. Cepstral mean subtracton (CMS) s a good compensaton technque for convolutve dstortons. Varance normalzaton conssts n normalzng the feature varance to one and n sgnal recognton to deal wth noses and channel msmatch. Normalzaton can be global or local. In the frst case, the mean and standard devaton are computed globally whle n the second case, they are computed on a wndow centered on the current tme. Feature extracton method LPC starts wth the assumpton that an audo sgnal s produced by a buzzer at the end of a tube, wth occasonal added hssng and poppng sounds. Although apparently crude, ths model s actually a close approxmaton to the realty of sgnal producton. LPC analyzes the sgnal by estmatng the formants, removng ther effects from the sgnal, and estmatng the ntensty and frequency of the remanng buzz. The process of removng the formants s called nverse flterng, and the remanng sgnal after the subtracton of the fltered modeled sgnal s called the resdue. The numbers, whch descrbe the ntensty and frequency of the buzz, the formants, and the resdue sgnal, can be stored or transmtted somewhere else. LPC syntheszes the sgnal by reversng the process: use the buzz parameters and the resdue to create a source sgnal, use the formants to create a flter, and run the source through the flter, resultng n audo. Steps:. Audo fles n mono or stereo recorded n natural or nsde lab, or taken from a standard data base 2. Extractng features by removng nose provded t s a fresh audo, otherwse for exstng audo, nose removal s not requred 3. Two phases have to be adopted: Tranng phase and testng phase 4. Tranng Phase: In ths phase, a set of representatve numbers are to be obtaned from an ntal set of numbers. BPA s used for learnng the audo fles 5. Testng phase: In ths phase, the representatve numbers obtaned n step 4 has to be used along wth the features obtaned from a test audo fle to obtan, an actvaton value. Ths value s compared wth a threshold and fnal decson s taken to retreve an audo fle or offer soluton to take further acton whch can be actvatng a system n a moble phone, etc. Fg.4 Flowchart for LPC analyss F. Results and Dscusson Cepstrum analyss s a nonlnear sgnal processng technque wth a varety of applcatons n areas such as speech and mage processng. The complex cepstrum for a sequence x s calculated by fndng the complex
5 228 IJCSNS Internatonal Journal of Computer Scence and Network Securty, VOL.8 No.0, October 2008 Natural logarthm of the Fourer transform of x, then the nverse Fourer transform of the resultng sequence. The complex cepstrum transformaton s central to the theory and applcaton of homomorphc systems, that s, systems that obey certan general rules of superposton. The real cepstrum of a sgnal x, sometmes called smply the cepstrum, s calculated by determnng the natural logarthm of magntude of the Fourer transform of x, then obtanng the nverse Fourer transform of the resultng sequence. It s dffcult to reconstruct the orgnal sequence from ts real cepstrum transformaton, as the real cepstrum s based only on the magntude of the Fourer transform for the sequence. Table, gves the cepstral coeffcents for 25 sample audo fles. Each row s a pattern used for tranng the ANN wth BPA. The topology of the ANN used s 6 6. In ths, 6 nodes n the nput layer, 6 nodes n the hdden layer and node n the output layer s used for proper tranng of ANN followed by audo mnng. TABLE CEPSTRAL FEATURES OBTAINED FROM SAMPLE AUDIO FILES S. No Inputs F F2 F3 F4 F5 F Tar get Lab elng F F6 are cepstral values. We can choose more than 6 values for an audo fle. Target labelng should be less than and greater than zero. When the number of audo fle ncreases, then more decmal values have to be ncorporated. III.CONCLUSION Audo of common brds and pet anmals have been recorded casually. The audo fle s sutably preprocessed followed by cepstral analyss and tranng ANN usng BPA. A set of fnal weghts wth 6 6 confguraton s obtaned wth 7350 teratons to reach.025 mean squared error rate. Ffty patterns have been used for tranng the ANN. Thrty patterns were used for testng (audo mnng). The results are close to 90% of mnng as the audo was recorded n open. The percentage of recognton and audo mnng accuracy has to be tested wth large number of new audo fles from the same set of brds and pet anmals IV. REFERENCES [] Le Lu and Hong-Jang Zhang, Content analyss for audo classfcaton and segmentaton., IEEE Transactons on Speech and Audo Processng, 0:504 56, October [2] T. Tolonen and M. Karjalanen, A computatonally effcent multptch analyss model, IEEE Transactons on Speech and Audo Processng, Vol. 8(No. 6):708 76, November [3] Haleh Vafae and Kenneth De Jong, Feature space transformaton usng genetc algorthms, IEEE Intellgent Systems, 3(2):57 65, March/Aprl 998. [4] Usama M. Fayyad, Data Mnng and Knowledge Dscovery: Makng Sense Out of Data, IEEE Expert, October 996, pp [5] Fortuna L, Grazan S, LoPrest M and Muscato G (992), Improvng back-propagaton learnng usng auxlary neural networks, Int. J of Cont., 55(4), pp
6 IJCSNS Internatonal Journal of Computer Scence and Network Securty, VOL.8 No.0, October [6] Lppmann R P (987) An ntroducton to computng wth neural nets, IEEE Trans. On Acoustcs, Speech and Sgnal Processng Magazne, V35, N4, pp [7] Yao Y L and Fang X D (993), Assessment of chp formng patterns wth tool wear progresson n machnng va neural networks, Int.J. Mach. Tools & Mfg, 33 (), pp [8] Hush D R and Horne B G (993), Progress n supervsed neural networks, IEEE Sgnal Proc. Mag., pp [9] Bernard Wdrow (990), 30 Years of adaptve neural networks: Perceptron, madalne and back-propagaton, Proc. of the IEEE, 8(9), pp [0] Hrose Y, Yamashta K Y and Hjya S (99), Back-propagaton algorthm whch vares the number of hdden unts, Neural Networks, 4, pp [] Smon Haykn, Neural Networks-A Comprehensve foundaton, 2 nd edton. V. ABOUT THE AUTHORS Processng. Surendra Shetty, s workng as Lecturer n MCA Department at NMAMIT, Ntte, Karkala Taluk, Udup Dstrct, Karnataka, INDIA. He has receved MCA Degree and pursung Ph.D degree n the area of Data Mnng at Mangalore Unversty. Hs man research nterest ncludes Data Mnng, Artfcal Intellgence & Sgnal Dr.K.K.Achary, s a Professor of Statstcs n the Dept. of Post-graduate Studes and Research n Statstcs, at Mangalore Unversty,Inda. He holds M.Sc. degree n Statststcs and Ph.D. n Appled Mathematcs from Indan Insttute of Scence, Bangalore, Inda. Hs current research nterests nclude Stochastc Models, Inventory Theory, Face Recognton Technques and Audo Data Mnng. Hs research papers have appeared n European Journal of Operatonal Research, Journal of Operatonal Research Socety, Opsearch, CCERO, Internatonal Journal of Informaton and Management Scences, Statstcal Methods and Amercan Journal of Mathematcs and Management.
The Development of Web Log Mining Based on Improve-K-Means Clustering Analysis
The Development of Web Log Mnng Based on Improve-K-Means Clusterng Analyss TngZhong Wang * College of Informaton Technology, Luoyang Normal Unversty, Luoyang, 471022, Chna [email protected] Abstract.
Forecasting the Demand of Emergency Supplies: Based on the CBR Theory and BP Neural Network
700 Proceedngs of the 8th Internatonal Conference on Innovaton & Management Forecastng the Demand of Emergency Supples: Based on the CBR Theory and BP Neural Network Fu Deqang, Lu Yun, L Changbng School
Lecture 2: Single Layer Perceptrons Kevin Swingler
Lecture 2: Sngle Layer Perceptrons Kevn Sngler [email protected] Recap: McCulloch-Ptts Neuron Ths vastly smplfed model of real neurons s also knon as a Threshold Logc Unt: W 2 A Y 3 n W n. A set of synapses
Forecasting the Direction and Strength of Stock Market Movement
Forecastng the Drecton and Strength of Stock Market Movement Jngwe Chen Mng Chen Nan Ye [email protected] [email protected] [email protected] Abstract - Stock market s one of the most complcated systems
Feature selection for intrusion detection. Slobodan Petrović NISlab, Gjøvik University College
Feature selecton for ntruson detecton Slobodan Petrovć NISlab, Gjøvk Unversty College Contents The feature selecton problem Intruson detecton Traffc features relevant for IDS The CFS measure The mrmr measure
Statistical Approach for Offline Handwritten Signature Verification
Journal of Computer Scence 4 (3): 181-185, 2008 ISSN 1549-3636 2008 Scence Publcatons Statstcal Approach for Offlne Handwrtten Sgnature Verfcaton 2 Debnath Bhattacharyya, 1 Samr Kumar Bandyopadhyay, 2
On-Line Fault Detection in Wind Turbine Transmission System using Adaptive Filter and Robust Statistical Features
On-Lne Fault Detecton n Wnd Turbne Transmsson System usng Adaptve Flter and Robust Statstcal Features Ruoyu L Remote Dagnostcs Center SKF USA Inc. 3443 N. Sam Houston Pkwy., Houston TX 77086 Emal: [email protected]
Offline Verification of Hand Written Signature using Adaptive Resonance Theory Net (Type-1)
Internatonal Journal of Sgnal Processng Systems Vol, No June 203 Offlne Verfcaton of Hand Wrtten Sgnature usng Adaptve Resonance Theory Net (Type-) Trtharaj Dash Veer Surendra Sa Unversty of Technology,
Time Delayed Independent Component Analysis for Data Quality Monitoring
IWSSIP 1-17th Internatonal Conference on Systems, Sgnals and Image Processng Tme Delayed Independent Component Analyss for Data Qualty Montorng José Márco Faer Sgnal Processng Laboratory, COE/Pol Federal
RESEARCH ON DUAL-SHAKER SINE VIBRATION CONTROL. Yaoqi FENG 1, Hanping QIU 1. China Academy of Space Technology (CAST) yaoqi.feng@yahoo.
ICSV4 Carns Australa 9- July, 007 RESEARCH ON DUAL-SHAKER SINE VIBRATION CONTROL Yaoq FENG, Hanpng QIU Dynamc Test Laboratory, BISEE Chna Academy of Space Technology (CAST) [email protected] Abstract
Frequency Selective IQ Phase and IQ Amplitude Imbalance Adjustments for OFDM Direct Conversion Transmitters
Frequency Selectve IQ Phase and IQ Ampltude Imbalance Adjustments for OFDM Drect Converson ransmtters Edmund Coersmeer, Ernst Zelnsk Noka, Meesmannstrasse 103, 44807 Bochum, Germany [email protected],
A Secure Password-Authenticated Key Agreement Using Smart Cards
A Secure Password-Authentcated Key Agreement Usng Smart Cards Ka Chan 1, Wen-Chung Kuo 2 and Jn-Chou Cheng 3 1 Department of Computer and Informaton Scence, R.O.C. Mltary Academy, Kaohsung 83059, Tawan,
Development of an intelligent system for tool wear monitoring applying neural networks
of Achevements n Materals and Manufacturng Engneerng VOLUME 14 ISSUE 1-2 January-February 2006 Development of an ntellgent system for tool wear montorng applyng neural networks A. Antć a, J. Hodolč a,
DEFINING %COMPLETE IN MICROSOFT PROJECT
CelersSystems DEFINING %COMPLETE IN MICROSOFT PROJECT PREPARED BY James E Aksel, PMP, PMI-SP, MVP For Addtonal Informaton about Earned Value Management Systems and reportng, please contact: CelersSystems,
A Hierarchical Anomaly Network Intrusion Detection System using Neural Network Classification
IDC IDC A Herarchcal Anomaly Network Intruson Detecton System usng Neural Network Classfcaton ZHENG ZHANG, JUN LI, C. N. MANIKOPOULOS, JAY JORGENSON and JOSE UCLES ECE Department, New Jersey Inst. of Tech.,
Biometric Signature Processing & Recognition Using Radial Basis Function Network
Bometrc Sgnature Processng & Recognton Usng Radal Bass Functon Network Ankt Chadha, Neha Satam, and Vbha Wal Abstract- Automatc recognton of sgnature s a challengng problem whch has receved much attenton
A DATA MINING APPLICATION IN A STUDENT DATABASE
JOURNAL OF AERONAUTICS AND SPACE TECHNOLOGIES JULY 005 VOLUME NUMBER (53-57) A DATA MINING APPLICATION IN A STUDENT DATABASE Şenol Zafer ERDOĞAN Maltepe Ünversty Faculty of Engneerng Büyükbakkalköy-Istanbul
A Crossplatform ECG Compression Library for Mobile HealthCare Services
A Crossplatform ECG Compresson Lbrary for Moble HealthCare Servces Alexander Borodn, Yulya Zavyalova Department of Computer Scence Petrozavodsk State Unversty Petrozavodsk, Russa {aborod, yzavyalo}@cs.petrsu.ru
Mining Multiple Large Data Sources
The Internatonal Arab Journal of Informaton Technology, Vol. 7, No. 3, July 2 24 Mnng Multple Large Data Sources Anmesh Adhkar, Pralhad Ramachandrarao 2, Bhanu Prasad 3, and Jhml Adhkar 4 Department of
Can Auto Liability Insurance Purchases Signal Risk Attitude?
Internatonal Journal of Busness and Economcs, 2011, Vol. 10, No. 2, 159-164 Can Auto Lablty Insurance Purchases Sgnal Rsk Atttude? Chu-Shu L Department of Internatonal Busness, Asa Unversty, Tawan Sheng-Chang
Improved SVM in Cloud Computing Information Mining
Internatonal Journal of Grd Dstrbuton Computng Vol.8, No.1 (015), pp.33-40 http://dx.do.org/10.1457/jgdc.015.8.1.04 Improved n Cloud Computng Informaton Mnng Lvshuhong (ZhengDe polytechnc college JangSu
A COLLABORATIVE TRADING MODEL BY SUPPORT VECTOR REGRESSION AND TS FUZZY RULE FOR DAILY STOCK TURNING POINTS DETECTION
A COLLABORATIVE TRADING MODEL BY SUPPORT VECTOR REGRESSION AND TS FUZZY RULE FOR DAILY STOCK TURNING POINTS DETECTION JHENG-LONG WU, PEI-CHANN CHANG, KAI-TING CHANG Department of Informaton Management,
APPLICATION OF PROBE DATA COLLECTED VIA INFRARED BEACONS TO TRAFFIC MANEGEMENT
APPLICATION OF PROBE DATA COLLECTED VIA INFRARED BEACONS TO TRAFFIC MANEGEMENT Toshhko Oda (1), Kochro Iwaoka (2) (1), (2) Infrastructure Systems Busness Unt, Panasonc System Networks Co., Ltd. Saedo-cho
An Evaluation of the Extended Logistic, Simple Logistic, and Gompertz Models for Forecasting Short Lifecycle Products and Services
An Evaluaton of the Extended Logstc, Smple Logstc, and Gompertz Models for Forecastng Short Lfecycle Products and Servces Charles V. Trappey a,1, Hsn-yng Wu b a Professor (Management Scence), Natonal Chao
Causal, Explanatory Forecasting. Analysis. Regression Analysis. Simple Linear Regression. Which is Independent? Forecasting
Causal, Explanatory Forecastng Assumes cause-and-effect relatonshp between system nputs and ts output Forecastng wth Regresson Analyss Rchard S. Barr Inputs System Cause + Effect Relatonshp The job of
What is Candidate Sampling
What s Canddate Samplng Say we have a multclass or mult label problem where each tranng example ( x, T ) conssts of a context x a small (mult)set of target classes T out of a large unverse L of possble
How To Understand The Results Of The German Meris Cloud And Water Vapour Product
Ttel: Project: Doc. No.: MERIS level 3 cloud and water vapour products MAPP MAPP-ATBD-ClWVL3 Issue: 1 Revson: 0 Date: 9.12.1998 Functon Name Organsaton Sgnature Date Author: Bennartz FUB Preusker FUB Schüller
Power-of-Two Policies for Single- Warehouse Multi-Retailer Inventory Systems with Order Frequency Discounts
Power-of-wo Polces for Sngle- Warehouse Mult-Retaler Inventory Systems wth Order Frequency Dscounts José A. Ventura Pennsylvana State Unversty (USA) Yale. Herer echnon Israel Insttute of echnology (Israel)
Gender Classification for Real-Time Audience Analysis System
Gender Classfcaton for Real-Tme Audence Analyss System Vladmr Khryashchev, Lev Shmaglt, Andrey Shemyakov, Anton Lebedev Yaroslavl State Unversty Yaroslavl, Russa [email protected], [email protected], [email protected],
8 Algorithm for Binary Searching in Trees
8 Algorthm for Bnary Searchng n Trees In ths secton we present our algorthm for bnary searchng n trees. A crucal observaton employed by the algorthm s that ths problem can be effcently solved when the
VRT012 User s guide V0.1. Address: Žirmūnų g. 27, Vilnius LT-09105, Phone: (370-5) 2127472, Fax: (370-5) 276 1380, Email: info@teltonika.
VRT012 User s gude V0.1 Thank you for purchasng our product. We hope ths user-frendly devce wll be helpful n realsng your deas and brngng comfort to your lfe. Please take few mnutes to read ths manual
Vision Mouse. Saurabh Sarkar a* University of Cincinnati, Cincinnati, USA ABSTRACT 1. INTRODUCTION
Vson Mouse Saurabh Sarkar a* a Unversty of Cncnnat, Cncnnat, USA ABSTRACT The report dscusses a vson based approach towards trackng of eyes and fngers. The report descrbes the process of locatng the possble
Implementation of Deutsch's Algorithm Using Mathcad
Implementaton of Deutsch's Algorthm Usng Mathcad Frank Roux The followng s a Mathcad mplementaton of Davd Deutsch's quantum computer prototype as presented on pages - n "Machnes, Logc and Quantum Physcs"
RELIABILITY, RISK AND AVAILABILITY ANLYSIS OF A CONTAINER GANTRY CRANE ABSTRACT
Kolowrock Krzysztof Joanna oszynska MODELLING ENVIRONMENT AND INFRATRUCTURE INFLUENCE ON RELIABILITY AND OPERATION RT&A # () (Vol.) March RELIABILITY RIK AND AVAILABILITY ANLYI OF A CONTAINER GANTRY CRANE
Calculation of Sampling Weights
Perre Foy Statstcs Canada 4 Calculaton of Samplng Weghts 4.1 OVERVIEW The basc sample desgn used n TIMSS Populatons 1 and 2 was a two-stage stratfed cluster desgn. 1 The frst stage conssted of a sample
INVESTIGATION OF VEHICULAR USERS FAIRNESS IN CDMA-HDR NETWORKS
21 22 September 2007, BULGARIA 119 Proceedngs of the Internatonal Conference on Informaton Technologes (InfoTech-2007) 21 st 22 nd September 2007, Bulgara vol. 2 INVESTIGATION OF VEHICULAR USERS FAIRNESS
Stochastic Protocol Modeling for Anomaly Based Network Intrusion Detection
Stochastc Protocol Modelng for Anomaly Based Network Intruson Detecton Juan M. Estevez-Tapador, Pedro Garca-Teodoro, and Jesus E. Daz-Verdejo Department of Electroncs and Computer Technology Unversty of
Intelligent Voice-Based Door Access Control System Using Adaptive-Network-based Fuzzy Inference Systems (ANFIS) for Building Security
Journal of Computer Scence 3 (5): 274-280, 2007 ISSN 1549-3636 2007 Scence Publcatons Intellgent Voce-Based Door Access Control System Usng Adaptve-Network-based Fuzzy Inference Systems (ANFIS) for Buldng
Bayesian Network Based Causal Relationship Identification and Funding Success Prediction in P2P Lending
Proceedngs of 2012 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 25 (2012) (2012) IACSIT Press, Sngapore Bayesan Network Based Causal Relatonshp Identfcaton and Fundng Success
A hybrid global optimization algorithm based on parallel chaos optimization and outlook algorithm
Avalable onlne www.ocpr.com Journal of Chemcal and Pharmaceutcal Research, 2014, 6(7):1884-1889 Research Artcle ISSN : 0975-7384 CODEN(USA) : JCPRC5 A hybrd global optmzaton algorthm based on parallel
Multiple-Period Attribution: Residuals and Compounding
Multple-Perod Attrbuton: Resduals and Compoundng Our revewer gave these authors full marks for dealng wth an ssue that performance measurers and vendors often regard as propretary nformaton. In 1994, Dens
An Enhanced Super-Resolution System with Improved Image Registration, Automatic Image Selection, and Image Enhancement
An Enhanced Super-Resoluton System wth Improved Image Regstraton, Automatc Image Selecton, and Image Enhancement Yu-Chuan Kuo ( ), Chen-Yu Chen ( ), and Chou-Shann Fuh ( ) Department of Computer Scence
Module 2 LOSSLESS IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur
Module LOSSLESS IMAGE COMPRESSION SYSTEMS Lesson 3 Lossless Compresson: Huffman Codng Instructonal Objectves At the end of ths lesson, the students should be able to:. Defne and measure source entropy..
ANALYZING THE RELATIONSHIPS BETWEEN QUALITY, TIME, AND COST IN PROJECT MANAGEMENT DECISION MAKING
ANALYZING THE RELATIONSHIPS BETWEEN QUALITY, TIME, AND COST IN PROJECT MANAGEMENT DECISION MAKING Matthew J. Lberatore, Department of Management and Operatons, Vllanova Unversty, Vllanova, PA 19085, 610-519-4390,
Face Verification Problem. Face Recognition Problem. Application: Access Control. Biometric Authentication. Face Verification (1:1 matching)
Face Recognton Problem Face Verfcaton Problem Face Verfcaton (1:1 matchng) Querymage face query Face Recognton (1:N matchng) database Applcaton: Access Control www.vsage.com www.vsoncs.com Bometrc Authentcaton
An Alternative Way to Measure Private Equity Performance
An Alternatve Way to Measure Prvate Equty Performance Peter Todd Parlux Investment Technology LLC Summary Internal Rate of Return (IRR) s probably the most common way to measure the performance of prvate
A Perfect QoS Routing Algorithm for Finding the Best Path for Dynamic Networks
A Perfect QoS Routng Algorthm for Fndng the Best Path for Dynamc Networks Hazem M. El-Bakry Faculty of Computer Scence & Informaton Systems, Mansoura Unversty, EGYPT [email protected] Nkos Mastoraks
1. Fundamentals of probability theory 2. Emergence of communication traffic 3. Stochastic & Markovian Processes (SP & MP)
6.3 / -- Communcaton Networks II (Görg) SS20 -- www.comnets.un-bremen.de Communcaton Networks II Contents. Fundamentals of probablty theory 2. Emergence of communcaton traffc 3. Stochastc & Markovan Processes
Credit Limit Optimization (CLO) for Credit Cards
Credt Lmt Optmzaton (CLO) for Credt Cards Vay S. Desa CSCC IX, Ednburgh September 8, 2005 Copyrght 2003, SAS Insttute Inc. All rghts reserved. SAS Propretary Agenda Background Tradtonal approaches to credt
THE APPLICATION OF DATA MINING TECHNIQUES AND MULTIPLE CLASSIFIERS TO MARKETING DECISION
Internatonal Journal of Electronc Busness Management, Vol. 3, No. 4, pp. 30-30 (2005) 30 THE APPLICATION OF DATA MINING TECHNIQUES AND MULTIPLE CLASSIFIERS TO MARKETING DECISION Yu-Mn Chang *, Yu-Cheh
Damage detection in composite laminates using coin-tap method
Damage detecton n composte lamnates usng con-tap method S.J. Km Korea Aerospace Research Insttute, 45 Eoeun-Dong, Youseong-Gu, 35-333 Daejeon, Republc of Korea [email protected] 45 The con-tap test has the
Design and Development of a Security Evaluation Platform Based on International Standards
Internatonal Journal of Informatcs Socety, VOL.5, NO.2 (203) 7-80 7 Desgn and Development of a Securty Evaluaton Platform Based on Internatonal Standards Yuj Takahash and Yoshm Teshgawara Graduate School
Luby s Alg. for Maximal Independent Sets using Pairwise Independence
Lecture Notes for Randomzed Algorthms Luby s Alg. for Maxmal Independent Sets usng Parwse Independence Last Updated by Erc Vgoda on February, 006 8. Maxmal Independent Sets For a graph G = (V, E), an ndependent
RequIn, a tool for fast web traffic inference
RequIn, a tool for fast web traffc nference Olver aul, Jean Etenne Kba GET/INT, LOR Department 9 rue Charles Fourer 90 Evry, France [email protected], [email protected] Abstract As networked
Simple Interest Loans (Section 5.1) :
Chapter 5 Fnance The frst part of ths revew wll explan the dfferent nterest and nvestment equatons you learned n secton 5.1 through 5.4 of your textbook and go through several examples. The second part
A Simple Approach to Clustering in Excel
A Smple Approach to Clusterng n Excel Aravnd H Center for Computatonal Engneerng and Networng Amrta Vshwa Vdyapeetham, Combatore, Inda C Rajgopal Center for Computatonal Engneerng and Networng Amrta Vshwa
Vehicle Detection and Tracking in Video from Moving Airborne Platform
Journal of Computatonal Informaton Systems 10: 12 (2014) 4965 4972 Avalable at http://www.jofcs.com Vehcle Detecton and Trackng n Vdeo from Movng Arborne Platform Lye ZHANG 1,2,, Hua WANG 3, L LI 2 1 School
ECE544NA Final Project: Robust Machine Learning Hardware via Classifier Ensemble
1 ECE544NA Fnal Project: Robust Machne Learnng Hardware va Classfer Ensemble Sa Zhang, [email protected] Dept. of Electr. & Comput. Eng., Unv. of Illnos at Urbana-Champagn, Urbana, IL, USA Abstract In
A Replication-Based and Fault Tolerant Allocation Algorithm for Cloud Computing
A Replcaton-Based and Fault Tolerant Allocaton Algorthm for Cloud Computng Tork Altameem Dept of Computer Scence, RCC, Kng Saud Unversty, PO Box: 28095 11437 Ryadh-Saud Araba Abstract The very large nfrastructure
Project Networks With Mixed-Time Constraints
Project Networs Wth Mxed-Tme Constrants L Caccetta and B Wattananon Western Australan Centre of Excellence n Industral Optmsaton (WACEIO) Curtn Unversty of Technology GPO Box U1987 Perth Western Australa
PAS: A Packet Accounting System to Limit the Effects of DoS & DDoS. Debish Fesehaye & Klara Naherstedt University of Illinois-Urbana Champaign
PAS: A Packet Accountng System to Lmt the Effects of DoS & DDoS Debsh Fesehaye & Klara Naherstedt Unversty of Illnos-Urbana Champagn DoS and DDoS DDoS attacks are ncreasng threats to our dgtal world. Exstng
THE DISTRIBUTION OF LOAN PORTFOLIO VALUE * Oldrich Alfons Vasicek
HE DISRIBUION OF LOAN PORFOLIO VALUE * Oldrch Alfons Vascek he amount of captal necessary to support a portfolo of debt securtes depends on the probablty dstrbuton of the portfolo loss. Consder a portfolo
A study on the ability of Support Vector Regression and Neural Networks to Forecast Basic Time Series Patterns
A study on the ablty of Support Vector Regresson and Neural Networks to Forecast Basc Tme Seres Patterns Sven F. Crone, Jose Guajardo 2, and Rchard Weber 2 Lancaster Unversty, Department of Management
benefit is 2, paid if the policyholder dies within the year, and probability of death within the year is ).
REVIEW OF RISK MANAGEMENT CONCEPTS LOSS DISTRIBUTIONS AND INSURANCE Loss and nsurance: When someone s subject to the rsk of ncurrng a fnancal loss, the loss s generally modeled usng a random varable or
An artificial Neural Network approach to monitor and diagnose multi-attribute quality control processes. S. T. A. Niaki*
Journal of Industral Engneerng Internatonal July 008, Vol. 4, No. 7, 04 Islamc Azad Unversty, South Tehran Branch An artfcal Neural Network approach to montor and dagnose multattrbute qualty control processes
MATHEMATICAL ENGINEERING TECHNICAL REPORTS. Sequential Optimizing Investing Strategy with Neural Networks
MATHEMATICAL ENGINEERING TECHNICAL REPORTS Sequental Optmzng Investng Strategy wth Neural Networks Ryo ADACHI and Akmch TAKEMURA METR 2010 03 February 2010 DEPARTMENT OF MATHEMATICAL INFORMATICS GRADUATE
A frequency decomposition time domain model of broadband frequency-dependent absorption: Model II
A frequenc decomposton tme doman model of broadband frequenc-dependent absorpton: Model II W. Chen Smula Research Laborator, P. O. Box. 134, 135 Lsaker, Norwa (1 Aprl ) (Proect collaborators: A. Bounam,
Gaining Insights to the Tea Industry of Sri Lanka using Data Mining
Proceedngs of the Internatonal MultConference of Engneers and Computer Scentsts 2008 Vol I Ganng Insghts to the Tea Industry of Sr Lanka usng Data Mnng H.C. Fernando, W. M. R Tssera, and R. I. Athauda
BUSINESS PROCESS PERFORMANCE MANAGEMENT USING BAYESIAN BELIEF NETWORK. 0688, [email protected]
Proceedngs of the 41st Internatonal Conference on Computers & Industral Engneerng BUSINESS PROCESS PERFORMANCE MANAGEMENT USING BAYESIAN BELIEF NETWORK Yeong-bn Mn 1, Yongwoo Shn 2, Km Jeehong 1, Dongsoo
Single and multiple stage classifiers implementing logistic discrimination
Sngle and multple stage classfers mplementng logstc dscrmnaton Hélo Radke Bttencourt 1 Dens Alter de Olvera Moraes 2 Vctor Haertel 2 1 Pontfíca Unversdade Católca do Ro Grande do Sul - PUCRS Av. Ipranga,
VoIP Playout Buffer Adjustment using Adaptive Estimation of Network Delays
VoIP Playout Buffer Adjustment usng Adaptve Estmaton of Network Delays Mroslaw Narbutt and Lam Murphy* Department of Computer Scence Unversty College Dubln, Belfeld, Dubln, IRELAND Abstract The poor qualty
A neuro-fuzzy collaborative filtering approach for Web recommendation. G. Castellano, A. M. Fanelli, and M. A. Torsello *
Internatonal Journal of Computatonal Scence 992-6669 (Prnt) 992-6677 (Onlne) Global Informaton Publsher 27, Vol., No., 27-39 A neuro-fuzzy collaboratve flterng approach for Web recommendaton G. Castellano,
A Passive Network Measurement-based Traffic Control Algorithm in Gateway of. P2P Systems
roceedngs of the 7th World Congress The Internatonal Federaton of Automatc Control A assve Network Measurement-based Traffc Control Algorthm n Gateway of 2 Systems Ybo Jang, Weje Chen, Janwe Zheng, Wanlang
Course outline. Financial Time Series Analysis. Overview. Data analysis. Predictive signal. Trading strategy
Fnancal Tme Seres Analyss Patrck McSharry [email protected] www.mcsharry.net Trnty Term 2014 Mathematcal Insttute Unversty of Oxford Course outlne 1. Data analyss, probablty, correlatons, vsualsaton
Performance Analysis and Coding Strategy of ECOC SVMs
Internatonal Journal of Grd and Dstrbuted Computng Vol.7, No. (04), pp.67-76 http://dx.do.org/0.457/jgdc.04.7..07 Performance Analyss and Codng Strategy of ECOC SVMs Zhgang Yan, and Yuanxuan Yang, School
An ILP Formulation for Task Mapping and Scheduling on Multi-core Architectures
An ILP Formulaton for Task Mappng and Schedulng on Mult-core Archtectures Yng Y, We Han, Xn Zhao, Ahmet T. Erdogan and Tughrul Arslan Unversty of Ednburgh, The Kng's Buldngs, Mayfeld Road, Ednburgh, EH9
A DYNAMIC CRASHING METHOD FOR PROJECT MANAGEMENT USING SIMULATION-BASED OPTIMIZATION. Michael E. Kuhl Radhamés A. Tolentino-Peña
Proceedngs of the 2008 Wnter Smulaton Conference S. J. Mason, R. R. Hll, L. Mönch, O. Rose, T. Jefferson, J. W. Fowler eds. A DYNAMIC CRASHING METHOD FOR PROJECT MANAGEMENT USING SIMULATION-BASED OPTIMIZATION
Large Scale Extreme Learning Machine using MapReduce
Large Scale Extreme Learnng Machne usng MapReduce Large Scale Extreme Learnng Machne usng MapReduce * L Dong, Pan Zhsong, 3 Deng Zhantao, 4 Zhang Yanyan Insttute of Command Automaton, PLA Unversty of Scence
8.5 UNITARY AND HERMITIAN MATRICES. The conjugate transpose of a complex matrix A, denoted by A*, is given by
6 CHAPTER 8 COMPLEX VECTOR SPACES 5. Fnd the kernel of the lnear transformaton gven n Exercse 5. In Exercses 55 and 56, fnd the mage of v, for the ndcated composton, where and are gven by the followng
Minimal Coding Network With Combinatorial Structure For Instantaneous Recovery From Edge Failures
Mnmal Codng Network Wth Combnatoral Structure For Instantaneous Recovery From Edge Falures Ashly Joseph 1, Mr.M.Sadsh Sendl 2, Dr.S.Karthk 3 1 Fnal Year ME CSE Student Department of Computer Scence Engneerng
Using Content-Based Filtering for Recommendation 1
Usng Content-Based Flterng for Recommendaton 1 Robn van Meteren 1 and Maarten van Someren 2 1 NetlnQ Group, Gerard Brandtstraat 26-28, 1054 JK, Amsterdam, The Netherlands, [email protected] 2 Unversty of
Lei Liu, Hua Yang Business School, Hunan University, Changsha, Hunan, P.R. China, 410082. Abstract
, pp.377-390 http://dx.do.org/10.14257/jsa.2016.10.4.34 Research on the Enterprse Performance Management Informaton System Development and Robustness Optmzaton based on Data Regresson Analyss and Mathematcal
Capacity-building and training
92 Toolkt to Combat Traffckng n Persons Tool 2.14 Capacty-buldng and tranng Overvew Ths tool provdes references to tranng programmes and materals. For more tranng materals, refer also to Tool 9.18. Capacty-buldng
320 The Internatonal Arab Journal of Informaton Technology, Vol. 5, No. 3, July 2008 Comparsons Between Data Clusterng Algorthms Osama Abu Abbas Computer Scence Department, Yarmouk Unversty, Jordan Abstract:
Efficient Project Portfolio as a tool for Enterprise Risk Management
Effcent Proect Portfolo as a tool for Enterprse Rsk Management Valentn O. Nkonov Ural State Techncal Unversty Growth Traectory Consultng Company January 5, 27 Effcent Proect Portfolo as a tool for Enterprse
Automated Network Performance Management and Monitoring via One-class Support Vector Machine
Automated Network Performance Management and Montorng va One-class Support Vector Machne R. Zhang, J. Jang, and S. Zhang Dgtal Meda & Systems Research Insttute, Unversty of Bradford, UK Abstract: In ths
Set. algorithms based. 1. Introduction. System Diagram. based. Exploration. 2. Index
ISSN (Prnt): 1694-0784 ISSN (Onlne): 1694-0814 www.ijcsi.org 236 IT outsourcng servce provder dynamc evaluaton model and algorthms based on Rough Set L Sh Sh 1,2 1 Internatonal School of Software, Wuhan
PSYCHOLOGICAL RESEARCH (PYC 304-C) Lecture 12
14 The Ch-squared dstrbuton PSYCHOLOGICAL RESEARCH (PYC 304-C) Lecture 1 If a normal varable X, havng mean µ and varance σ, s standardsed, the new varable Z has a mean 0 and varance 1. When ths standardsed
