Transactions on Information and Communications Technologies vol WIT Press, ISSN

Size: px
Start display at page:

Download "Transactions on Information and Communications Technologies vol WIT Press, ISSN"

Transcription

1 A comparison of methods for customer classification Maria Celia S.Lopes, Myrian C.A.Costa & Nelson F.F.Ebecken COPPE/Federal University ofrio de Janeiro Caixa Postal CEP Rio de Janeiro - R7- Brazil Phone.: (55 21) Fax.: (55 21) Abstract This work presents a comparison of current methods used for classification problems. The solution of two typical applications related to Customer Classification for Business Applications are considered: the first proposed in the Second International Competition of Data Analysis by Intelligent Techniques [6] and a higher dimensionality case. 1 Introduction Recently, databases with hundred of fields and tables and million of records are being treated with data mining tools. A very large number of records in the database and a very large number of fields, generate a high dimensionality problem. This creates difficulties in terms of increasing the size of the search space for model induction and generates spurious patterns related to irrelevant variables. This paper is organized as follows: section 2 presents a description of the employed methods; in section 3, the problem 1 and results are briefly described, section 4 deals with a more complex problem; finally, section 5 presents some conclusions.

2 2 Data Mining Methods M I -DECISION TREE - Personal Computer Implementation [7] This induction algorithm is considered to be binary, once it creates a two way branch at every split in the tree. The selection of the attribute to split on at every stage in tree building is done according to the information contents of each attribute in terms of classifying the outcome groups. The most informative attribute is selected at every branching point. For discrete attributes, the value groups are split between the two branches so as to maximize the information content to the attribute. For numeric attributes the two way split is based on a numeric threshold which is derived to maximize the information content to the attribute. When the outcome is numeric the standard deviation of the data filtering to both branches are used as the basis for selecting the best attribute and the best threshold. One of the parameters that have to be specified before induction process commences is the Minimum Examples in a Branch. This figure gives the induction algorithm a criteria for stopping the creation of new branches from any given point in the tree if the number of data samples filtering to that point falls bellow this limit. This limit provides defence against noise in the data. In effect it will only allow branches to be developed from an acceptable number of records. Normally this figure is set depending on the total number of records and the level of noise in the data. MII -DECISION TREE - Workstation Implementation [8] This Decision tree algorithm builds a classification model in the form of a binary tree that can be interpreted visually or by reading rules in if-then format. The model starts at the root node and follows a path determined by the attribute test until a leaf node is encountered. Each leaf node has a label assigned that represents the classification of the record. The split used to create the binary tree employs a breadth-first tree-growing technique and depends on the type of the attribute considered. If the attribute is numeric, the splits are of the form: 334

3 A <v, where A is the attribute and v is a numeric value for this attribute. If the attribute is discrete, the method considers splits of the type: A E S\ where S(A) are the set of possible values for attributed and S' c:s. M III -NEURAL INDUCTION- Workstation Implementation [4] The Neural Induction algorithm employs a back-propagation neural network, with heuristic search for the best network architecture, to produce one trained network and the sensitivity analysis of the attributes as outputs. The back-propagation is a general purpose supervised learning algorithm. The sensitivity analysis will show how fields contribute to the classification. The resulting classification model can then be used to predict the classes of new attribute values as well as a ranked list of fields relevant to the classification. In this particular implementation the classification is based on the value of one classfield.the complete algorithm has four parts: a) Normalization, where the data are examined to determine how the values are translated into format required by the input. b) Selection of architectures, that chooses several network configurations with different numbers of hidden units based on the number of inputs and outputs units. c) Training of architectures, using the back-propagation algorithm. d) Choose the best network architecture and overall training, based on a score calculated from desired accuracy, error limits and complexity of networks M IV - MULTILAYERED PERCEPTRON/Exaustive Network Search [3] This method utilizes the Multi-Layer Perceptron (MLP) model and a constructive approach to build networks in conjunction with an adaptive gradient learning rule. The algorithm of network construction is characterized by the addition of hidden units one or a few at a time. 335

4 Construction is stopped when performance on an independent test set shows no further improvement. The method follows the steps listed above in order to build a neural network model: a) analyzing and converting data into a form suitable for the network inputs; b) attributes selection, that utilizes a genetic algorithm to search for good sets of input attributes and for each possible set, a logistic regression or a neural network is trained and used to rank the subsets of inputs; c) network construction and train, using a method of network construction with an adaptive gradient learning rule. M V - MULTILAYERED PERCEPTRON/Genetic Algorithm [5] Multi-layered Perceptron (MLP) is a nonparametric architecture. Used with the backpropagation algorithm it is capable of generating smooth nonlinear mappings between input and output variables. The multilayered perceptron is considered a type of neural network. Both of these terms come from the fact that this architecture was originally proposed as a model for neural biological processes. However, in this approach, this vantage point is ignored, viewing MLP simply as a useful architecture for nonparametric modeling. Output Layer Hidden Layer Intput Figure 1 - A MLP T network T Intput Layer 336

5 As shown in figure 1, an MLP can be viewed as an interconnected network made up of nodes that can be thought of as simple computacional elements. The nodes are arranged into one or more layers. The first layer is the input layer, the final layer is the output layer, and the other layers inbetween are called hidden layers. The output of a node in a hidden layer is used as an input to the nodes of the next layer. Each hidden node outputs the value obtained from applying a sigmoidal function of a weighted sum of its inputs. In classification, the output nodes also apply a sigmoidal function. In estimation, output nodes simply output weighted sums of their inputs. A separate weight exists for each connection in the network (i.e., between each pair of nodes in adjoining layers). These are the weights used by nodes to weigh the inputs they are summing, and they constitute the free parameters to be tuned by data. Backpropagation is one of many errorminimazing functions which tunes these weights to generate the desired maping. The error function used is usually the mean-squared error (MSB) over a data set. M VI - GENERALIZED REGRESSION NEURAL NETWORK [2] GRNN are a 4 layer feed forward neural network that accepts discrete and/or continuous valued inputs and generates discrete or continues valued outputs. GRNN also memorize the training records by storing the input and output variables in the network itself. Once stored, as new records are presented, the GRNN looks at the difference between the current record and all the stored records and performs what can be thought of as an interpolation and generates an estimated output based on the history stored in the network. Unlike back propagation, which attempts to create a mathematical formula that generates outputs based on inputs, the GRNN is performing an interpolated estimate within its previous experience. GRNN uses a recall factor called "sigma" to adjust the acuity (degree of discrimination) of neural network's response. Sigma is inversely related to acuity thus a low sigma generates a highly discriminate response while a high sigma generates a more generalized response. GRNN are sometimes a good substitute for Back Propagation, within the GRNN's constraints (covered below). They can be used for regression and time series types of applications. 337

6 Advantages of GRNN: Fast Training Speed for modest training sets Sometimes higher accuracy than Back Propagation Disadvantages of GRNN: Large networks with large training Data Sets Slower performance on "recall" with large networks Do not handle trending inputs and outputs well Difficult to determine proper "recall factor" (sigma). GRNN memorizes the data in one pass. This reduces training greatly as compared to iterative techniques like Back Propagation. Depending on the nature of the data, it can be much more accurate than Back Propagation. This can mean some outstanding advantages. Since GRNN memorize the training data, large training sets (records) can create large networks. There is one first hidden layer node created for each output (plus 1). Thus, for training records and 2 predicting outputs using 5 inputs, this ends up with a 5 input, hidden, 3 hidden, 2 output network. If Back Propagation is used, a much smaller network might be found. Since these networks can get large, depending on the application, their "recall time", the time to pass new data through them, can be slow. Also, since they are comparing current records to a previous history and are not creating a mathematical "formula" relating inputs and outputs like Back Propagation, they do not handle data that is outside the range they were trained on very well. Additionally, the knowing what a proper "sigma" factor to use can be problematic. It depends on how discriminating the output is needed. This issue is reduced quite a bit by automatically optimizing sigma in the algorithm as the networks are built. M VII - TIME DELAY NEURAL NETWORK [1] TDNN is a more general form of Back Propagation. It employs the Back Propagation technique for setting weights between neurons, but they also 338

7 consider time inherently in the structure of its architecture. It can be viewed as a back propagation network where there are multiple connections from the input neurons to the output neurons. Each of these connections looks back over time and sets its weights for each connection to minimize Mean Squared Error (MSB) of the overall network. Figure 2 depicts such a network. Input(M) Output(t) Optional Recurrent Outputs Intput Layer Hidden Layer Output Layer Figure 2 - A TDNN network The network shown in Figure 2 is a Time Delay Neural Network with two (2) connections from each input neuron to each hidden neuron, and one connection from the hidden neuron to the output neuron. In a TDNN, each connection is set to a specific data interval back in time with the first connection set at the current time, (current record) and the second connection set to one period ago (third connection to 2 periods ago, etc.) This look back is performed by providing each neuron with memory, so that it can remember previous layer outputs for N periods of time. Thus, TDNN can be thought of a as a back propagation network with fixed time delays back N periods of time, exactly like lagging your inputs by N periods of time. The one major difference is that TDNN also does this with hidden neurons output too, thus seeing, remembering and using "features" in your data over time. TDNN, in this implementation, has also the option of being recurrent. This means that the network architectures can be set to take the last output of the network and use it as an input. These inputs also include the TDNN time-based look back ability, thus it can look back over the history of the neural outputs for extended time span features of the data. The Look Back feature of TDNN networks makes them particularly appropriate for time series applications where histories of input variables 339

8 are used to produce predictions into the future. TDNN network can also be used for time based classification and diagnostics where histories of inputs are used to identify the existence of some condition. Advantages of TDNN networks: No input lagging required, just load the data in time sequential order and run Uses familiar back propagation techniques Much higher accuracy on some problem types than back propagation Disadvantages of TDNN networks: Time delays consume usable records (it takes records to "pre-load" a network just like with recurrent networks and much like records lost when lagging inputs manually) More free parameters requires more data More free parameters means relatively slower training 3 Problem 1 - Description Customer Classification for Business Applications [6] The objective of database marketing is the discovery of inhomogeneous information of the customer's personal and demographic background as well as the products the customer already uses. The whole information extracted from the databases are used in order to select as exactly as possible those persons from a database who show the greatest potential to actually buy an advertised product or to get information about the customers requirements. The data records included personal data, such as age, gender and job as well as detailed information about the customers' product utilization. 18 different products were considered, ranging from checking and savings accounts to investment plans and securities. Additionally, a classification of their place of residence was available for each person in the database. Altogether about 100 data fields were available for each customer in the database. 27 from the total of 100 fields were selected as input data. The whole set of data contains 300 data sets, including personal data and information about the customers product utilization and the information 340

9 whether the customer has ordered a special product during a marketing campaign. Parts of the data are encoded to protect the personal data without a loss of information. The training data are in ASCII format. Table 1: Structure of the data of problem 1 Feature 1 pd,j Feature 2 Pd,,2 Feature j Pd,,i Feature 27 Pd,,27 pd,so,i Pdi50,2 Pdisoj Pdi50,27 Explanation: pdy: personal dataj of data set i (i=l,..., 300; j=l,..., 27) Feature : customer personal data; Feature : product utilization data (0 or 1) RESULTS FOR PROBLEM 1 M I -DECISION TREE - Personal Computer Implementation The Building of this tree was complete, presenting the folowing results: Number of classes = 2 General Accuracy for class 1: 93.30%. General Accuracy for class 0: 93.30%. Overall Accuracy: 93.3% M II -DECISION TREE - Workstation Implementation The Building of this tree was complete, presenting the folowing results: Number of classes = 2 Errors =64(21,33%) Predicted Class Total = 150 Total =150 Total =

10 M III -NEURAL INDUCTION- Workstation Implementation The Training of this network was complete, presenting the folowing results: General Accuracy for class 1: 83.30%. General Accuracy for class 0: 82.60%. Actual Unknown M IV - MULTILAYERED PERCEPTRON/Genetic Algorithm The Training of this network was complete, presenting the folowing results: Mean Squared Error on training set: Min. Mean Squared Error on test set: This network is a Multilayered Perceptron neural network, employing 9 inputs and 19 hidden layers. There were 1 output neuron. M V - MULTILAYERED PERCEPTRON/Genetic Algorithm The Training of this network was complete, presenting the folowing results: Mean Squared Error on training set: This network is a Multilayered Perceptron neural network. All the columns in the datafile were used. M VI - GENERALIZED REGRESSION NEURAL NETWORK The Training of this network was complete, presenting the folowing results: Mean Squared Error on training set: Min. Mean Squared Error on test set: This network is a Generalized Regression neural network, employing 27 inputs and 2 hidden layers. The second hidden layer used a summation transfer function. There were 1 output neurons using a direct transfer function. All the columns in the datafile were used. 342

11 M VII - TIME DELAY NEURAL NETWORK The Training of this network was complete, presenting the folowing results: Accuracy on training set: 98.00%. Max. accuracy on test set: 84.00%. This network is a Time Delay neural network, employing 24 inputs and 2 hidden layers. Thefirsthidden layer had 4 Tanh 2 Linear neurons with 3 connections. The second hidden layer had 2 Logistic 2 Linear neurons with 7 connections. There were 1 output neurons using a linear transfer function and 1 connections each. The following columns in the datafile were used: Cl, C2, C3, C5, C7, C8, C9, CIO, Cll, PI, P2, P3, P5, P6, P7, P8, P9, P10, P11,P12,P13,P14,P15,P16 COMPARISON OF RESULTS In figure 3, thre was an attempt to show the results of the described methods. + Class Class M-ll A Class M-III X Class M-IV X Class M-V Class M-VI + ClassM-VII Figure 3 - Outputs Comparison of the results of Methods II, HI, IV, V, VI, VII 343

12 4 Problem 2 - Description Problem 2 has extracted data from a real-world insurance database. A data set of 32 attributes and registers was collected, from a huge database containing information of insured persons and companies. This set of data has typical properties like fragmentation, varying data quality, irregular data value coding, missing values, noise, etc. which make the application of data mining a challenge. The complexity and dimensionality of this problem brings some discussions about algorithms and their results. The database describes relations among customers, insurance contracts and components of insurance tariffs. Each customer can play roles in certain insurance policies and an insurance contract can have several components, each of which is related with a tariff role of the respective customer. Each policy concerns a certain product and tariff components are bound to dedicated insurance tariffs. The 32 attributes are distributed as follows: - 9 attributes describes the customer, like sex, birth date, marital status, etc.; - 11 attributes has policy information, like type of contract, status of contract, modus of payment, etc.; - 12 attributes specify the tariff components, like insured benefits and regular premium. Several methods were been used to discover the relationship among attributes, but only two of them were considered since the results are reasonable. The first result utilizes the method 2 to classify data. The obtained decision tree has 43 levels of depth. This run identified 3 classes with 3624 registers (11,32%) erroneously classified. Figure 4 shows the comparison of the predicted and the target classification of an excerpt of a hundred points. 344

13 . Target classification Q Predicted classification 0, Figure 4 -Target x Predicted Classification The second result uses method 3. The obtained neural network classified the data with 63,11% of accuracy. The Figure 5 presents the resulting and the desired classification of a hundred outputs. 2,5 2 1,5-1 0,5 0.Desired classification & Resulting Classification Figure 5 - Desired x Resulting Classification 5 Conclusion In this work, two considerable difficult problems were chosen to examine the main characteristics of usual customer classification problem methods. 345

14 These methods are selected from those most commonly employed today's software. No special effort was made in the amalgamation of the raw data. The simply consideration of the 28 attributes resulted on a hard task. Problem 1 (28att x 300 records) could be easily analised on personal computer platforms. In this case, the Multilayered Perceptron/Exaustive Network Search corresponded to the more accurated solution. In Problem 2 (32att x records), the solutions could only be obtained on workstation implementations. Naturally, The induction solution have shown na excellent computer performance. In terms of accuracy, again the neural net strategy yielded the best results. As a conclusion, the focused data can be considered as benchmarks to the research of mining methods implemented to scale the dimensions of very large databases. REFERENCES [1] Masters, T., Signal and Image Processing -with Neural Networks, John Wiley and Sons, Inc, USA, pp , [2] Masters, T., Advanced Algorithms for Neural Networks, John Wiley and Sons, Inc, New York, pp , [3] Michie, D., Spiegelhalter, D.J. and Taylor, C.C., Machine Learning, Neural and Statistical Classification, Ellis Horwood Limited, [4] Arbib, M. A., The Handbook of Brain Theory and Neural Networks, The MIT Press, Massachusetts, pp , [5] Kennedy, R.L., Lee, Y., Van Roy, B., Reed, C.D. and Lippmann, R.P., Solving Data Mining Problems through Pattern Recognition, Prentice Hall, USA, 1997 [6] ERUDIT'98 - Second International Competition of Data Analysys by Intelligent Techiniques - European Network of Excellence for Uncertainty 346

15 Modelling - Sep [7] Quinlan, J.R., C4.5 Programs for Machine Learning, Morgan Kaufmann Publishers, California, 1993 [8] Agrawal, R., Mannila, H., Srikant, R., Toivonen, H., Verkano, A.I., Fast Discovery of Association Rules, Chapter 12, Advances in knowledge Discovery and Data Mining, eds. Fayyad, U.M., Shapiro, G.R., Smyth P. and Uthurusamy, R., The MIT Press, California, pp ,

Neural Networks. CE-725: Statistical Pattern Recognition Sharif University of Technology Spring Soleymani

Neural Networks. CE-725: Statistical Pattern Recognition Sharif University of Technology Spring Soleymani Neural Networks CE-725: Statistical Pattern Recognition Sharif University of Technology Spring 2013 Soleymani Outline Biological and artificial neural networks Feed-forward neural networks Single layer

More information

Analytical model A structure and process for analyzing a dataset. For example, a decision tree is a model for the classification of a dataset.

Analytical model A structure and process for analyzing a dataset. For example, a decision tree is a model for the classification of a dataset. Glossary of data mining terms: Accuracy Accuracy is an important factor in assessing the success of data mining. When applied to data, accuracy refers to the rate of correct values in the data. When applied

More information

CS 4510/9010 Applied Machine Learning. Neural Nets. Paula Matuszek Fall copyright Paula Matuszek 2016

CS 4510/9010 Applied Machine Learning. Neural Nets. Paula Matuszek Fall copyright Paula Matuszek 2016 CS 4510/9010 Applied Machine Learning 1 Neural Nets Paula Matuszek Fall 2016 Neural Nets, the very short version 2 A neural net consists of layers of nodes, or neurons, each of which has an activation

More information

A Systematic Overview of Data Mining Algorithms

A Systematic Overview of Data Mining Algorithms A Systematic Overview of Data Mining Algorithms 1 Data Mining Algorithm A well-defined procedure that takes data as input and produces output as models or patterns well-defined: precisely encoded as a

More information

Learning. Learning agents Inductive learning. Neural Networks. Different Learning Scenarios Evaluation

Learning. Learning agents Inductive learning. Neural Networks. Different Learning Scenarios Evaluation Learning Learning agents Inductive learning Different Learning Scenarios Evaluation Slides based on Slides by Russell/Norvig, Ronald Williams, and Torsten Reil Material from Russell & Norvig, chapters

More information

Data Mining. Neural Networks

Data Mining. Neural Networks Data Mining Neural Networks Goals for this Unit Basic understanding of Neural Networks and how they work Ability to use Neural Networks to solve real problems Understand when neural networks may be most

More information

A Systematic Overview of Data Mining Algorithms. Sargur Srihari University at Buffalo The State University of New York

A Systematic Overview of Data Mining Algorithms. Sargur Srihari University at Buffalo The State University of New York A Systematic Overview of Data Mining Algorithms Sargur Srihari University at Buffalo The State University of New York 1 Topics Data Mining Algorithm Definition Example of CART Classification Iris, Wine

More information

Performance Analysis of Data Mining Classification Techniques

Performance Analysis of Data Mining Classification Techniques Performance Analysis of Data Mining Classification Techniques Tejas Mehta 1, Dr. Dhaval Kathiriya 2 Ph.D. Student, School of Computer Science, Dr. Babasaheb Ambedkar Open University, Gujarat, India 1 Principal

More information

Data Mining. Part 2. Data Understanding and Preparation. 2.4 Data Transformation. Spring Instructor: Dr. Masoud Yaghini. Data Transformation

Data Mining. Part 2. Data Understanding and Preparation. 2.4 Data Transformation. Spring Instructor: Dr. Masoud Yaghini. Data Transformation Data Mining Part 2. Data Understanding and Preparation 2.4 Spring 2010 Instructor: Dr. Masoud Yaghini Outline Introduction Normalization Attribute Construction Aggregation Attribute Subset Selection Discretization

More information

Decision Trees Dr. G. Bharadwaja Kumar VIT Chennai

Decision Trees Dr. G. Bharadwaja Kumar VIT Chennai Decision Trees Decision Tree Decision Trees (DTs) are a nonparametric supervised learning method used for classification and regression. The goal is to create a model that predicts the value of a target

More information

LECTURE NOTES Professor Anita Wasilewska NEURAL NETWORKS

LECTURE NOTES Professor Anita Wasilewska NEURAL NETWORKS LECTURE NOTES Professor Anita Wasilewska NEURAL NETWORKS Neural Networks Classifier Introduction INPUT: classification data, i.e. it contains an classification (class) attribute. WE also say that the class

More information

Cse634 DATA MINING TEST REVIEW. Professor Anita Wasilewska Computer Science Department Stony Brook University

Cse634 DATA MINING TEST REVIEW. Professor Anita Wasilewska Computer Science Department Stony Brook University Cse634 DATA MINING TEST REVIEW Professor Anita Wasilewska Computer Science Department Stony Brook University Preprocessing stage Preprocessing: includes all the operations that have to be performed before

More information

CS6220: DATA MINING TECHNIQUES

CS6220: DATA MINING TECHNIQUES CS6220: DATA MINING TECHNIQUES Image Data: Classification via Neural Networks Instructor: Yizhou Sun yzsun@ccs.neu.edu November 19, 2015 Methods to Learn Classification Clustering Frequent Pattern Mining

More information

ISSN: (Online) Volume 3, Issue 9, September 2015 International Journal of Advance Research in Computer Science and Management Studies

ISSN: (Online) Volume 3, Issue 9, September 2015 International Journal of Advance Research in Computer Science and Management Studies ISSN: 2321-7782 (Online) Volume 3, Issue 9, September 2015 International Journal of Advance Research in Computer Science and Management Studies Research Article / Survey Paper / Case Study Available online

More information

Classification Lecture Notes cse352. Neural Networks. Professor Anita Wasilewska

Classification Lecture Notes cse352. Neural Networks. Professor Anita Wasilewska Classification Lecture Notes cse352 Neural Networks Professor Anita Wasilewska Neural Networks Classification Introduction INPUT: classification data, i.e. it contains an classification (class) attribute

More information

Natural Language Processing CS 6320 Lecture 6 Neural Language Models. Instructor: Sanda Harabagiu

Natural Language Processing CS 6320 Lecture 6 Neural Language Models. Instructor: Sanda Harabagiu Natural Language Processing CS 6320 Lecture 6 Neural Language Models Instructor: Sanda Harabagiu In this lecture We shall cover: Deep Neural Models for Natural Language Processing Introduce Feed Forward

More information

Character Recognition Using Convolutional Neural Networks

Character Recognition Using Convolutional Neural Networks Character Recognition Using Convolutional Neural Networks David Bouchain Seminar Statistical Learning Theory University of Ulm, Germany Institute for Neural Information Processing Winter 2006/2007 Abstract

More information

CMPT 882 Week 3 Summary

CMPT 882 Week 3 Summary CMPT 882 Week 3 Summary! Artificial Neural Networks (ANNs) are networks of interconnected simple units that are based on a greatly simplified model of the brain. ANNs are useful learning tools by being

More information

Enhancing Forecasting Performance of Naïve-Bayes Classifiers with Discretization Techniques

Enhancing Forecasting Performance of Naïve-Bayes Classifiers with Discretization Techniques 24 Enhancing Forecasting Performance of Naïve-Bayes Classifiers with Discretization Techniques Enhancing Forecasting Performance of Naïve-Bayes Classifiers with Discretization Techniques Ruxandra PETRE

More information

International Journal of Scientific Research & Engineering Trends Volume 4, Issue 6, Nov-Dec-2018, ISSN (Online): X

International Journal of Scientific Research & Engineering Trends Volume 4, Issue 6, Nov-Dec-2018, ISSN (Online): X Analysis about Classification Techniques on Categorical Data in Data Mining Assistant Professor P. Meena Department of Computer Science Adhiyaman Arts and Science College for Women Uthangarai, Krishnagiri,

More information

For Monday. Read chapter 18, sections Homework:

For Monday. Read chapter 18, sections Homework: For Monday Read chapter 18, sections 10-12 The material in section 8 and 9 is interesting, but we won t take time to cover it this semester Homework: Chapter 18, exercise 25 a-b Program 4 Model Neuron

More information

Applying Supervised Learning

Applying Supervised Learning Applying Supervised Learning When to Consider Supervised Learning A supervised learning algorithm takes a known set of input data (the training set) and known responses to the data (output), and trains

More information

A Neural Network Model Of Insurance Customer Ratings

A Neural Network Model Of Insurance Customer Ratings A Neural Network Model Of Insurance Customer Ratings Jan Jantzen 1 Abstract Given a set of data on customers the engineering problem in this study is to model the data and classify customers

More information

Dynamic Analysis of Structures Using Neural Networks

Dynamic Analysis of Structures Using Neural Networks Dynamic Analysis of Structures Using Neural Networks Alireza Lavaei Academic member, Islamic Azad University, Boroujerd Branch, Iran Alireza Lohrasbi Academic member, Islamic Azad University, Boroujerd

More information

Data Mining and Analytics

Data Mining and Analytics Data Mining and Analytics Aik Choon Tan, Ph.D. Associate Professor of Bioinformatics Division of Medical Oncology Department of Medicine aikchoon.tan@ucdenver.edu 9/22/2017 http://tanlab.ucdenver.edu/labhomepage/teaching/bsbt6111/

More information

Supervised Learning in Neural Networks (Part 2)

Supervised Learning in Neural Networks (Part 2) Supervised Learning in Neural Networks (Part 2) Multilayer neural networks (back-propagation training algorithm) The input signals are propagated in a forward direction on a layer-bylayer basis. Learning

More information

2. Neural network basics

2. Neural network basics 2. Neural network basics Next commonalities among different neural networks are discussed in order to get started and show which structural parts or concepts appear in almost all networks. It is presented

More information

Random Forest A. Fornaser

Random Forest A. Fornaser Random Forest A. Fornaser alberto.fornaser@unitn.it Sources Lecture 15: decision trees, information theory and random forests, Dr. Richard E. Turner Trees and Random Forests, Adele Cutler, Utah State University

More information

Ensemble methods in machine learning. Example. Neural networks. Neural networks

Ensemble methods in machine learning. Example. Neural networks. Neural networks Ensemble methods in machine learning Bootstrap aggregating (bagging) train an ensemble of models based on randomly resampled versions of the training set, then take a majority vote Example What if you

More information

Knowledge Discovery and Data Mining. Neural Nets. A simple NN as a Mathematical Formula. Notes. Lecture 13 - Neural Nets. Tom Kelsey.

Knowledge Discovery and Data Mining. Neural Nets. A simple NN as a Mathematical Formula. Notes. Lecture 13 - Neural Nets. Tom Kelsey. Knowledge Discovery and Data Mining Lecture 13 - Neural Nets Tom Kelsey School of Computer Science University of St Andrews http://tom.home.cs.st-andrews.ac.uk twk@st-andrews.ac.uk Tom Kelsey ID5059-13-NN

More information

AMOL MUKUND LONDHE, DR.CHELPA LINGAM

AMOL MUKUND LONDHE, DR.CHELPA LINGAM International Journal of Advances in Applied Science and Engineering (IJAEAS) ISSN (P): 2348-1811; ISSN (E): 2348-182X Vol. 2, Issue 4, Dec 2015, 53-58 IIST COMPARATIVE ANALYSIS OF ANN WITH TRADITIONAL

More information

Perceptrons and Backpropagation. Fabio Zachert Cognitive Modelling WiSe 2014/15

Perceptrons and Backpropagation. Fabio Zachert Cognitive Modelling WiSe 2014/15 Perceptrons and Backpropagation Fabio Zachert Cognitive Modelling WiSe 2014/15 Content History Mathematical View of Perceptrons Network Structures Gradient Descent Backpropagation (Single-Layer-, Multilayer-Networks)

More information

Knowledge Discovery and Data Mining

Knowledge Discovery and Data Mining Knowledge Discovery and Data Mining Lecture 13 - Neural Nets Tom Kelsey School of Computer Science University of St Andrews http://tom.home.cs.st-andrews.ac.uk twk@st-andrews.ac.uk Tom Kelsey ID5059-13-NN

More information

Dynamic Load Balancing of Unstructured Computations in Decision Tree Classifiers

Dynamic Load Balancing of Unstructured Computations in Decision Tree Classifiers Dynamic Load Balancing of Unstructured Computations in Decision Tree Classifiers A. Srivastava E. Han V. Kumar V. Singh Information Technology Lab Dept. of Computer Science Information Technology Lab Hitachi

More information

Hybrid PSO-SA algorithm for training a Neural Network for Classification

Hybrid PSO-SA algorithm for training a Neural Network for Classification Hybrid PSO-SA algorithm for training a Neural Network for Classification Sriram G. Sanjeevi 1, A. Naga Nikhila 2,Thaseem Khan 3 and G. Sumathi 4 1 Associate Professor, Dept. of CSE, National Institute

More information

MIT 801. Machine Learning I. [Presented by Anna Bosman] 16 February 2018

MIT 801. Machine Learning I. [Presented by Anna Bosman] 16 February 2018 MIT 801 [Presented by Anna Bosman] 16 February 2018 Machine Learning What is machine learning? Artificial Intelligence? Yes as we know it. What is intelligence? The ability to acquire and apply knowledge

More information

Review on Methods of Selecting Number of Hidden Nodes in Artificial Neural Network

Review on Methods of Selecting Number of Hidden Nodes in Artificial Neural Network Available Online at www.ijcsmc.com International Journal of Computer Science and Mobile Computing A Monthly Journal of Computer Science and Information Technology IJCSMC, Vol. 3, Issue. 11, November 2014,

More information

Recitation Supplement: Creating a Neural Network for Classification SAS EM December 2, 2002

Recitation Supplement: Creating a Neural Network for Classification SAS EM December 2, 2002 Recitation Supplement: Creating a Neural Network for Classification SAS EM December 2, 2002 Introduction Neural networks are flexible nonlinear models that can be used for regression and classification

More information

Back propagation Algorithm:

Back propagation Algorithm: Network Neural: A neural network is a class of computing system. They are created from very simple processing nodes formed into a network. They are inspired by the way that biological systems such as the

More information

Introduction to Data Mining

Introduction to Data Mining Introduction to JULY 2011 Afsaneh Yazdani What motivated? Wide availability of huge amounts of data and the imminent need for turning such data into useful information and knowledge What motivated? Data

More information

Machine Learning 13. week

Machine Learning 13. week Machine Learning 13. week Deep Learning Convolutional Neural Network Recurrent Neural Network 1 Why Deep Learning is so Popular? 1. Increase in the amount of data Thanks to the Internet, huge amount of

More information

Assignment 2. Classification and Regression using Linear Networks, Multilayer Perceptron Networks, and Radial Basis Functions

Assignment 2. Classification and Regression using Linear Networks, Multilayer Perceptron Networks, and Radial Basis Functions ENEE 739Q: STATISTICAL AND NEURAL PATTERN RECOGNITION Spring 2002 Assignment 2 Classification and Regression using Linear Networks, Multilayer Perceptron Networks, and Radial Basis Functions Aravind Sundaresan

More information

Data Mining. 3.2 Decision Tree Classifier. Fall Instructor: Dr. Masoud Yaghini. Chapter 5: Decision Tree Classifier

Data Mining. 3.2 Decision Tree Classifier. Fall Instructor: Dr. Masoud Yaghini. Chapter 5: Decision Tree Classifier Data Mining 3.2 Decision Tree Classifier Fall 2008 Instructor: Dr. Masoud Yaghini Outline Introduction Basic Algorithm for Decision Tree Induction Attribute Selection Measures Information Gain Gain Ratio

More information

Lecture #11: The Perceptron

Lecture #11: The Perceptron Lecture #11: The Perceptron Mat Kallada STAT2450 - Introduction to Data Mining Outline for Today Welcome back! Assignment 3 The Perceptron Learning Method Perceptron Learning Rule Assignment 3 Will be

More information

Neural Network Neurons

Neural Network Neurons Neural Networks Neural Network Neurons 1 Receives n inputs (plus a bias term) Multiplies each input by its weight Applies activation function to the sum of results Outputs result Activation Functions Given

More information

Rough Set Approach to Unsupervised Neural Network based Pattern Classifier

Rough Set Approach to Unsupervised Neural Network based Pattern Classifier Rough Set Approach to Unsupervised Neural based Pattern Classifier Ashwin Kothari, Member IAENG, Avinash Keskar, Shreesha Srinath, and Rakesh Chalsani Abstract Early Convergence, input feature space with

More information

Logical Rhythm - Class 3. August 27, 2018

Logical Rhythm - Class 3. August 27, 2018 Logical Rhythm - Class 3 August 27, 2018 In this Class Neural Networks (Intro To Deep Learning) Decision Trees Ensemble Methods(Random Forest) Hyperparameter Optimisation and Bias Variance Tradeoff Biological

More information

11/14/2010 Intelligent Systems and Soft Computing 1

11/14/2010 Intelligent Systems and Soft Computing 1 Lecture 7 Artificial neural networks: Supervised learning Introduction, or how the brain works The neuron as a simple computing element The perceptron Multilayer neural networks Accelerated learning in

More information

Liquefaction Analysis in 3D based on Neural Network Algorithm

Liquefaction Analysis in 3D based on Neural Network Algorithm Liquefaction Analysis in 3D based on Neural Network Algorithm M. Tolon Istanbul Technical University, Turkey D. Ural Istanbul Technical University, Turkey SUMMARY: Simplified techniques based on in situ

More information

Constructively Learning a Near-Minimal Neural Network Architecture

Constructively Learning a Near-Minimal Neural Network Architecture Constructively Learning a Near-Minimal Neural Network Architecture Justin Fletcher and Zoran ObradoviC Abetract- Rather than iteratively manually examining a variety of pre-specified architectures, a constructive

More information

Cluster analysis of 3D seismic data for oil and gas exploration

Cluster analysis of 3D seismic data for oil and gas exploration Data Mining VII: Data, Text and Web Mining and their Business Applications 63 Cluster analysis of 3D seismic data for oil and gas exploration D. R. S. Moraes, R. P. Espíndola, A. G. Evsukoff & N. F. F.

More information

CSC 578 Neural Networks and Deep Learning

CSC 578 Neural Networks and Deep Learning CSC 578 Neural Networks and Deep Learning Fall 2018/19 7. Recurrent Neural Networks (Some figures adapted from NNDL book) 1 Recurrent Neural Networks 1. Recurrent Neural Networks (RNNs) 2. RNN Training

More information

Machine Learning. Deep Learning. Eric Xing (and Pengtao Xie) , Fall Lecture 8, October 6, Eric CMU,

Machine Learning. Deep Learning. Eric Xing (and Pengtao Xie) , Fall Lecture 8, October 6, Eric CMU, Machine Learning 10-701, Fall 2015 Deep Learning Eric Xing (and Pengtao Xie) Lecture 8, October 6, 2015 Eric Xing @ CMU, 2015 1 A perennial challenge in computer vision: feature engineering SIFT Spin image

More information

COMPUTATIONAL INTELLIGENCE

COMPUTATIONAL INTELLIGENCE COMPUTATIONAL INTELLIGENCE Fundamentals Adrian Horzyk Preface Before we can proceed to discuss specific complex methods we have to introduce basic concepts, principles, and models of computational intelligence

More information

Notes on Multilayer, Feedforward Neural Networks

Notes on Multilayer, Feedforward Neural Networks Notes on Multilayer, Feedforward Neural Networks CS425/528: Machine Learning Fall 2012 Prepared by: Lynne E. Parker [Material in these notes was gleaned from various sources, including E. Alpaydin s book

More information

Climate Precipitation Prediction by Neural Network

Climate Precipitation Prediction by Neural Network Journal of Mathematics and System Science 5 (205) 207-23 doi: 0.7265/259-529/205.05.005 D DAVID PUBLISHING Juliana Aparecida Anochi, Haroldo Fraga de Campos Velho 2. Applied Computing Graduate Program,

More information

CLASSIFICATION WITH RADIAL BASIS AND PROBABILISTIC NEURAL NETWORKS

CLASSIFICATION WITH RADIAL BASIS AND PROBABILISTIC NEURAL NETWORKS CLASSIFICATION WITH RADIAL BASIS AND PROBABILISTIC NEURAL NETWORKS CHAPTER 4 CLASSIFICATION WITH RADIAL BASIS AND PROBABILISTIC NEURAL NETWORKS 4.1 Introduction Optical character recognition is one of

More information

Multilayer Feed-forward networks

Multilayer Feed-forward networks Multi Feed-forward networks 1. Computational models of McCulloch and Pitts proposed a binary threshold unit as a computational model for artificial neuron. This first type of neuron has been generalized

More information

Assignment # 5. Farrukh Jabeen Due Date: November 2, Neural Networks: Backpropation

Assignment # 5. Farrukh Jabeen Due Date: November 2, Neural Networks: Backpropation Farrukh Jabeen Due Date: November 2, 2009. Neural Networks: Backpropation Assignment # 5 The "Backpropagation" method is one of the most popular methods of "learning" by a neural network. Read the class

More information

Decision Tree CE-717 : Machine Learning Sharif University of Technology

Decision Tree CE-717 : Machine Learning Sharif University of Technology Decision Tree CE-717 : Machine Learning Sharif University of Technology M. Soleymani Fall 2012 Some slides have been adapted from: Prof. Tom Mitchell Decision tree Approximating functions of usually discrete

More information

An Integer Recurrent Artificial Neural Network for Classifying Feature Vectors

An Integer Recurrent Artificial Neural Network for Classifying Feature Vectors An Integer Recurrent Artificial Neural Network for Classifying Feature Vectors Roelof K Brouwer PEng, PhD University College of the Cariboo, Canada Abstract: The main contribution of this report is the

More information

CS570: Introduction to Data Mining

CS570: Introduction to Data Mining CS570: Introduction to Data Mining Classification Advanced Reading: Chapter 8 & 9 Han, Chapters 4 & 5 Tan Anca Doloc-Mihu, Ph.D. Slides courtesy of Li Xiong, Ph.D., 2011 Han, Kamber & Pei. Data Mining.

More information

4.12 Generalization. In back-propagation learning, as many training examples as possible are typically used.

4.12 Generalization. In back-propagation learning, as many training examples as possible are typically used. 1 4.12 Generalization In back-propagation learning, as many training examples as possible are typically used. It is hoped that the network so designed generalizes well. A network generalizes well when

More information

DEEP LEARNING REVIEW. Yann LeCun, Yoshua Bengio & Geoffrey Hinton Nature Presented by Divya Chitimalla

DEEP LEARNING REVIEW. Yann LeCun, Yoshua Bengio & Geoffrey Hinton Nature Presented by Divya Chitimalla DEEP LEARNING REVIEW Yann LeCun, Yoshua Bengio & Geoffrey Hinton Nature 2015 -Presented by Divya Chitimalla What is deep learning Deep learning allows computational models that are composed of multiple

More information

International Journal of Electrical and Computer Engineering 4: Application of Neural Network in User Authentication for Smart Home System

International Journal of Electrical and Computer Engineering 4: Application of Neural Network in User Authentication for Smart Home System Application of Neural Network in User Authentication for Smart Home System A. Joseph, D.B.L. Bong, and D.A.A. Mat Abstract Security has been an important issue and concern in the smart home systems. Smart

More information

Adaptive Building of Decision Trees by Reinforcement Learning

Adaptive Building of Decision Trees by Reinforcement Learning Proceedings of the 7th WSEAS International Conference on Applied Informatics and Communications, Athens, Greece, August 24-26, 2007 34 Adaptive Building of Decision Trees by Reinforcement Learning MIRCEA

More information

CSE 626: Data mining. Instructor: Sargur N. Srihari. Phone: , ext. 113

CSE 626: Data mining. Instructor: Sargur N. Srihari.   Phone: , ext. 113 CSE 626: Data mining Instructor: Sargur N. Srihari E-mail: srihari@cedar.buffalo.edu Phone: 645-6164, ext. 113 1 What is Data Mining? Different perspectives: CSE, Business, IT As a field of research in

More information

Data Mining: Concepts and Techniques. Chapter 9 Classification: Support Vector Machines. Support Vector Machines (SVMs)

Data Mining: Concepts and Techniques. Chapter 9 Classification: Support Vector Machines. Support Vector Machines (SVMs) Data Mining: Concepts and Techniques Chapter 9 Classification: Support Vector Machines 1 Support Vector Machines (SVMs) SVMs are a set of related supervised learning methods used for classification Based

More information

Neural Networks for Machine Learning. Lecture 15a From Principal Components Analysis to Autoencoders

Neural Networks for Machine Learning. Lecture 15a From Principal Components Analysis to Autoencoders Neural Networks for Machine Learning Lecture 15a From Principal Components Analysis to Autoencoders Geoffrey Hinton Nitish Srivastava, Kevin Swersky Tijmen Tieleman Abdel-rahman Mohamed Principal Components

More information

Decision Trees Oct

Decision Trees Oct Decision Trees Oct - 7-2009 Previously We learned two different classifiers Perceptron: LTU KNN: complex decision boundary If you are a novice in this field, given a classification application, are these

More information

7. Decision or classification trees

7. Decision or classification trees 7. Decision or classification trees Next we are going to consider a rather different approach from those presented so far to machine learning that use one of the most common and important data structure,

More information

Research Article International Journals of Advanced Research in Computer Science and Software Engineering ISSN: X (Volume-7, Issue-6)

Research Article International Journals of Advanced Research in Computer Science and Software Engineering ISSN: X (Volume-7, Issue-6) International Journals of Advanced Research in Computer Science and Software Engineering Research Article June 17 Artificial Neural Network in Classification A Comparison Dr. J. Jegathesh Amalraj * Assistant

More information

Lecture 2 Notes. Outline. Neural Networks. The Big Idea. Architecture. Instructors: Parth Shah, Riju Pahwa

Lecture 2 Notes. Outline. Neural Networks. The Big Idea. Architecture. Instructors: Parth Shah, Riju Pahwa Instructors: Parth Shah, Riju Pahwa Lecture 2 Notes Outline 1. Neural Networks The Big Idea Architecture SGD and Backpropagation 2. Convolutional Neural Networks Intuition Architecture 3. Recurrent Neural

More information

Argha Roy* Dept. of CSE Netaji Subhash Engg. College West Bengal, India.

Argha Roy* Dept. of CSE Netaji Subhash Engg. College West Bengal, India. Volume 3, Issue 3, March 2013 ISSN: 2277 128X International Journal of Advanced Research in Computer Science and Software Engineering Research Paper Available online at: www.ijarcsse.com Training Artificial

More information

Pattern Recognition. Kjell Elenius. Speech, Music and Hearing KTH. March 29, 2007 Speech recognition

Pattern Recognition. Kjell Elenius. Speech, Music and Hearing KTH. March 29, 2007 Speech recognition Pattern Recognition Kjell Elenius Speech, Music and Hearing KTH March 29, 2007 Speech recognition 2007 1 Ch 4. Pattern Recognition 1(3) Bayes Decision Theory Minimum-Error-Rate Decision Rules Discriminant

More information

MLPQNA-LEMON Multi Layer Perceptron neural network trained by Quasi Newton or Levenberg-Marquardt optimization algorithms

MLPQNA-LEMON Multi Layer Perceptron neural network trained by Quasi Newton or Levenberg-Marquardt optimization algorithms MLPQNA-LEMON Multi Layer Perceptron neural network trained by Quasi Newton or Levenberg-Marquardt optimization algorithms 1 Introduction In supervised Machine Learning (ML) we have a set of data points

More information

Simple Model Selection Cross Validation Regularization Neural Networks

Simple Model Selection Cross Validation Regularization Neural Networks Neural Nets: Many possible refs e.g., Mitchell Chapter 4 Simple Model Selection Cross Validation Regularization Neural Networks Machine Learning 10701/15781 Carlos Guestrin Carnegie Mellon University February

More information

Elena Marchiori Free University Amsterdam, Faculty of Science, Department of Mathematics and Computer Science, Amsterdam, The Netherlands

Elena Marchiori Free University Amsterdam, Faculty of Science, Department of Mathematics and Computer Science, Amsterdam, The Netherlands DATA MINING Elena Marchiori Free University Amsterdam, Faculty of Science, Department of Mathematics and Computer Science, Amsterdam, The Netherlands Keywords: Data mining, knowledge discovery in databases,

More information

Implementation of a Library for Artificial Neural Networks in C

Implementation of a Library for Artificial Neural Networks in C Implementation of a Library for Artificial Neural Networks in C Jack Breese TJHSST Computer Systems Lab 2007-2008 June 10, 2008 1 Abstract In modern computing, there are several approaches to pattern recognition

More information

COMP 465: Data Mining Classification Basics

COMP 465: Data Mining Classification Basics Supervised vs. Unsupervised Learning COMP 465: Data Mining Classification Basics Slides Adapted From : Jiawei Han, Micheline Kamber & Jian Pei Data Mining: Concepts and Techniques, 3 rd ed. Supervised

More information

Allstate Insurance Claims Severity: A Machine Learning Approach

Allstate Insurance Claims Severity: A Machine Learning Approach Allstate Insurance Claims Severity: A Machine Learning Approach Rajeeva Gaur SUNet ID: rajeevag Jeff Pickelman SUNet ID: pattern Hongyi Wang SUNet ID: hongyiw I. INTRODUCTION The insurance industry has

More information

Data Mining. Covering algorithms. Covering approach At each stage you identify a rule that covers some of instances. Fig. 4.

Data Mining. Covering algorithms. Covering approach At each stage you identify a rule that covers some of instances. Fig. 4. Data Mining Chapter 4. Algorithms: The Basic Methods (Covering algorithm, Association rule, Linear models, Instance-based learning, Clustering) 1 Covering approach At each stage you identify a rule that

More information

Neuro-Fuzzy Inverse Forward Models

Neuro-Fuzzy Inverse Forward Models CS9 Autumn Neuro-Fuzzy Inverse Forward Models Brian Highfill Stanford University Department of Computer Science Abstract- Internal cognitive models are useful methods for the implementation of motor control

More information

Computational Intelligence Meets the NetFlix Prize

Computational Intelligence Meets the NetFlix Prize Computational Intelligence Meets the NetFlix Prize Ryan J. Meuth, Paul Robinette, Donald C. Wunsch II Abstract The NetFlix Prize is a research contest that will award $1 Million to the first group to improve

More information

Supervised Learning with Neural Networks. We now look at how an agent might learn to solve a general problem by seeing examples.

Supervised Learning with Neural Networks. We now look at how an agent might learn to solve a general problem by seeing examples. Supervised Learning with Neural Networks We now look at how an agent might learn to solve a general problem by seeing examples. Aims: to present an outline of supervised learning as part of AI; to introduce

More information

Artificial Neural Networks MLP, RBF & GMDH

Artificial Neural Networks MLP, RBF & GMDH Artificial Neural Networks MLP, RBF & GMDH Jan Drchal drchajan@fel.cvut.cz Computational Intelligence Group Department of Computer Science and Engineering Faculty of Electrical Engineering Czech Technical

More information

Neural Networks and Deep Learning

Neural Networks and Deep Learning Neural Networks and Deep Learning Example Learning Problem Example Learning Problem Celebrity Faces in the Wild Machine Learning Pipeline Raw data Feature extract. Feature computation Inference: prediction,

More information

Integrating Logistic Regression with Knowledge Discovery Systems

Integrating Logistic Regression with Knowledge Discovery Systems Association for Information Systems AIS Electronic Library (AISeL) AMCIS 1997 Proceedings Americas Conference on Information Systems (AMCIS) 8-15-1997 Integrating Logistic Regression with Knowledge Discovery

More information

Ubiquitous Computing and Communication Journal (ISSN )

Ubiquitous Computing and Communication Journal (ISSN ) A STRATEGY TO COMPROMISE HANDWRITTEN DOCUMENTS PROCESSING AND RETRIEVING USING ASSOCIATION RULES MINING Prof. Dr. Alaa H. AL-Hamami, Amman Arab University for Graduate Studies, Amman, Jordan, 2011. Alaa_hamami@yahoo.com

More information

Univariate and Multivariate Decision Trees

Univariate and Multivariate Decision Trees Univariate and Multivariate Decision Trees Olcay Taner Yıldız and Ethem Alpaydın Department of Computer Engineering Boğaziçi University İstanbul 80815 Turkey Abstract. Univariate decision trees at each

More information

Data Analysis and Solver Plugins for KSpread USER S MANUAL. Tomasz Maliszewski

Data Analysis and Solver Plugins for KSpread USER S MANUAL. Tomasz Maliszewski Data Analysis and Solver Plugins for KSpread USER S MANUAL Tomasz Maliszewski tmaliszewski@wp.pl Table of Content CHAPTER 1: INTRODUCTION... 3 1.1. ABOUT DATA ANALYSIS PLUGIN... 3 1.3. ABOUT SOLVER PLUGIN...

More information

Image Compression: An Artificial Neural Network Approach

Image Compression: An Artificial Neural Network Approach Image Compression: An Artificial Neural Network Approach Anjana B 1, Mrs Shreeja R 2 1 Department of Computer Science and Engineering, Calicut University, Kuttippuram 2 Department of Computer Science and

More information

Fuzzy Partitioning with FID3.1

Fuzzy Partitioning with FID3.1 Fuzzy Partitioning with FID3.1 Cezary Z. Janikow Dept. of Mathematics and Computer Science University of Missouri St. Louis St. Louis, Missouri 63121 janikow@umsl.edu Maciej Fajfer Institute of Computing

More information

The Data Mining usage in Production System Management

The Data Mining usage in Production System Management The Data Mining usage in Production System Management Pavel Vazan, Pavol Tanuska, Michal Kebisek Abstract The paper gives the pilot results of the project that is oriented on the use of data mining techniques

More information

Big Data Methods. Chapter 5: Machine learning. Big Data Methods, Chapter 5, Slide 1

Big Data Methods. Chapter 5: Machine learning. Big Data Methods, Chapter 5, Slide 1 Big Data Methods Chapter 5: Machine learning Big Data Methods, Chapter 5, Slide 1 5.1 Introduction to machine learning What is machine learning? Concerned with the study and development of algorithms that

More information

A *69>H>N6 #DJGC6A DG C<>C::G>C<,8>:C8:H /DA 'D 2:6G, ()-"&"3 -"(' ( +-" " " % '.+ % ' -0(+$,

A *69>H>N6 #DJGC6A DG C<>C::G>C<,8>:C8:H /DA 'D 2:6G, ()-&3 -(' ( +-   % '.+ % ' -0(+$, The structure is a very important aspect in neural network design, it is not only impossible to determine an optimal structure for a given problem, it is even impossible to prove that a given structure

More information

Nearest neighbor classification DSE 220

Nearest neighbor classification DSE 220 Nearest neighbor classification DSE 220 Decision Trees Target variable Label Dependent variable Output space Person ID Age Gender Income Balance Mortgag e payment 123213 32 F 25000 32000 Y 17824 49 M 12000-3000

More information

Credit card Fraud Detection using Predictive Modeling: a Review

Credit card Fraud Detection using Predictive Modeling: a Review February 207 IJIRT Volume 3 Issue 9 ISSN: 2396002 Credit card Fraud Detection using Predictive Modeling: a Review Varre.Perantalu, K. BhargavKiran 2 PG Scholar, CSE, Vishnu Institute of Technology, Bhimavaram,

More information

Motivation. Problem: With our linear methods, we can train the weights but not the basis functions: Activator Trainable weight. Fixed basis function

Motivation. Problem: With our linear methods, we can train the weights but not the basis functions: Activator Trainable weight. Fixed basis function Neural Networks Motivation Problem: With our linear methods, we can train the weights but not the basis functions: Activator Trainable weight Fixed basis function Flashback: Linear regression Flashback:

More information

Combining Models to Improve Classifier Accuracy and Robustness 1

Combining Models to Improve Classifier Accuracy and Robustness 1 Combining Models to Improve Classifier Accuracy and Robustness 1 Dean W. Abbott Abbott Consulting P.O. Box 22536 San Diego, CA 92192-2536 USA Email: dean@abbott-consulting.com Abstract Recent years have

More information