DOCSLIB.ORG

Development of a Simple Graphical Interface Based Software for Machine Learning and Data Visualization

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text

Load more

International Journal of Recent Technology and Engineering (IJRTE) ISSN: 2277-3878, Volume-8 Issue-2, July 2019 Development of a Simple Graphical Interface Based Software for Machine Learning and Data Visualization Ogunleye G.O., Fashoto S.G, Daramola C.Y., Ogundele L.A., Ojewumi T.O., Adewole Timilehin Abstract: Machine learning has become one of the foremost Clustering algorithms such as Mean Shift and KMeans; and techniques used for extracting knowledge from large amounts of deep learning algorithms such as the Artificial Neural data. The programming expertise required to implement machine Network. The technical know-how required to implement learning algorithms has led to the rise of software products that the various algorithms available in machine learning makes simplify the process. Many of these systems however, have it a challenge for non-programmers to be able to make use sacrificed simplicity as they evolved and included more features. In this study, a machine learning software with a simple of them. Therefore, it is necessary to provide a graphical graphical user interface was developed with a special focus on user interface (GUI) based application which hides the enhancing usability. The system made use of basic graphical implementation details of these algorithms from the user and interface elements such as buttons and textboxes. Comparison of also provide visualizations, therefore making it easy for non- the system with other similar open-source tools revealed that the technical users to use. There are various existing open developed system showed an improvement in usability over the source software packages that provide graphical user other tools. interface functionalities for machine learning tasks such as Keywords: GUI, Software, Machine Learning, data WEKA, RapidMiner, KNIME, and Orange. visualization The Waikato Environment for Knowledge Analysis, WEKA [25] is one such system. It provides easy access to I. INTRODUCTION machine learning tasks such as data preprocessing, Due to the increase in data generating devices and classification, regression, clustering, visualization and systems, there has been an exponential increase in the feature selection with both GUI and command line interface. amount of data stored in electronic format in the past few RapidMiner Studio [4] is a similar platform developed by decades[23]. As the volume of data being generated is RapidMiner. It has both proprietary and open source increasing, so is the amount of effort required for human versions. It also provides an integrated environment for easy beings to manually process them. This led to the preprocessing, machine learning, and predictive analytics. introduction of machine learning, an attempt to automate the However, some of its features are limited in the open source data analysis and processing procedures. version. Machine learning can be defined as the art of giving KNIME (Konstanz Information Miner) is a data analytics, machines the ability to detect meaningful patterns in data reporting and integration platform that uses modular data and make decisions based on these patterns without being pipelining to implement its machine learning components. explicitly programmed [1]. This involves giving the While these applications have their advantages, the steep machine a training data, generating a model from this learning curve associated with them due to their large training data and then using this model to predict future amount of functionalities and also the large amount of data. This is very useful for tasks that are too complex for system memory they occupy has created the need to produce explicit programming and tasks that require adaptivity. a simpler software focused on accommodating beginners. The field of Machine learning is one of the fastest The primary purpose of this paper is to develop an growing fields of computer science, with a wide range of application which simplifies machine learning tasks by applications. These applications include, but are not limited abstracting them into simple GUI operations. The software to: speech recognition (Deng and Li[3], fraud detection [1], presented in this paper will provide users without the recommender systems [5], anti-spam filters, medical knowledge needed to write programs the opportunity to diagnosis[3], bioinformatics [2], and computer vision. perform basic machine learning and visualization tasks by There are many algorithms available to perform these simplifying these tasks into GUI operations. It will also help tasks. Examples are classification algorithms such as K- experienced machine learning programmers who wish to nearest neighbors, Support Vector Classifier and Naïve quickly build machine learning models for small tasks to Bayes classifier; Regression algorithms such as Linear save time. The algorithms to be considered in the proposed Regression and Support Vector Regression; software are limited to: i. K-nearest neighbors, ii.Support Revised Manuscript Received on July 15, 2019. vector classifier, iii. Linear regression, iv. Ridge regression, Ogunleye G.O., Department of Computer Science, Federal University, v. Lasso regression, vi. Support vector regression, vii. K- Oye-Ekiti, Ekiti State, Nigeria. Fashoto S.G, Department of Computer Science, University of Means, viii. Mean shift, ix. Artificial neural network. Swaziland, Kwaluseni, Swaziland Daramola C.Y., Department of Computer Science, Federal University, II. RELATED WORKS Oye-Ekiti, Ekiti State, Nigeria Ogundele L.A., Department of Computer Science, College of 2.1 Overview Education, Ilesa, Osun State, Nigeria Ojewumi T.O., Department of Computer Science, Redeemer’s As computer technology is University, Ede, Osun State, Nigeria advancing, more people are Adewole Timilehin, Department of Computer Science, Federal University, Oye-Ekiti, Ekiti State, Nigeria making use of computer Published By: Retrieval Number: B3426078219/19©BEIESP Blue Eyes Intelligence Engineering DOI: 10.35940/ijrte.B3426.078219 3770 & Sciences Publication Development of a Simple Graphical Interface Based Software for Machine Learning and Data Visualization systems for various tasks. This implies that more people al.[14]) and Scikit-learn [10] are examples of general without much knowledge of programming need to perform purpose tools while KNIME [17] is a tool with focus on tasks that require expert programming knowledge. As a computations in Chemistry but with general purpose result, many efforts have been made to solve this problem applications. by creating software tools that can significantly simplify 2.2.4 Target users these tasks [13]. Tools made for non-technical users tend to be more The tools created for machine learning have varied intuitive, visually-oriented and less technical. Such tools in many aspects such as type (library, platform, framework), also feature reduced functionalities in order to avoid feature intended users (beginners, experts), interface (CLI, GUI, clutter. This ensures that simplicity, ease of use and API), domain (business, computational biology, statistics, learnability are maintained. general), licensing (open-source, proprietary) and Tools made for experienced users on the other hand usually functionality among many others. In the rest of this chapter, favor functionality and flexibility more than ease of use and some of these characteristics are discussed and some learnability. existing tools are reviewed. 2.3 Existing Machine Learning Tools 2.2 Features of Machine Learning Tools In the past couple of decades, machine learning has become 2.2.1 Software type a popular method for performing tasks that require the Programming languages are the lower level tools for extraction of information from datasets [6]. Several software machine learning. The choice of programming language tools have been developed to make machine learning used for a machine learning project may have a huge impact programming easier and faster. They are in the form of: on the results. This can be due to factors such as ease of use, i) Programming languages, speed, scalability, extensibility, availability of libraries, data ii) Libraries/Packages, structures and other programming structures [8]. Some of iii) Graphical User Interface-based the commonly used programming languages are Python, R, applications, platforms and toolkits, C++ and Java. iv) Combinations of any of these. Libraries and packages are collections of facilities 2.3.1 Programming Languages such as functions and methods that can be called by the user. Python Libraries usually provide only discrete facilities, that is, Python is one of the most popular programming language their facilities are specialized for specific tasks. for machine learning because of its simple syntax, its Platforms and toolkits provide more functionality modular architecture, and its huge repository of libraries. It than libraries, often enough to complete whole projects. can be challenging though for beginners to master the They can contain their own integrated development language because of its steep learning curve. environments. Their features are loosely coupled, enabling R Statistical Programming Language the user to tie them together in various ways [20]. R is a similar language designed mainly for statistical 2.2.2 User Interface computing, data mining and machine learning. This domain- User interface design is one of the central issues on the specific nature of the language together with its huge usability of a software having a huge impact on the repository of libraries positioned
Recommended publications
  • Quick Tour of Machine Learning (機器 )

    Quick Tour of Machine Learning (機器 )

    Quick Tour of Machine Learning (_hx習速遊) Hsuan-Tien Lin (林軒0) [email protected] Department of Computer Science & Information Engineering National Taiwan University (國Ë台c'xÇ訊工程û) Ç料科x愛}者t會û列;動, 2015/12/12 Hsuan-Tien Lin (NTU CSIE) Quick Tour of Machine Learning 0/128 Learning from Data Disclaimer just super-condensed and shuffled version of • • my co-authored textbook “Learning from Data: A Short Course” • my two NTU-Coursera Mandarin-teaching ML Massive Open Online Courses • “Machine Learning Foundations”: www.coursera.org/course/ntumlone • “Machine Learning Techniques”: www.coursera.org/course/ntumltwo —impossible to be complete, with most math details removed live interaction is important • goal: help you begin your journey with ML Hsuan-Tien Lin (NTU CSIE) Quick Tour of Machine Learning 1/128 Learning from Data Roadmap Learning from Data What is Machine Learning Components of Machine Learning Types of Machine Learning Step-by-step Machine Learning Hsuan-Tien Lin (NTU CSIE) Quick Tour of Machine Learning 2/128 Learning from Data What is Machine Learning Learning from Data :: What is Machine Learning Hsuan-Tien Lin (NTU CSIE) Quick Tour of Machine Learning 3/128 learning: machine learning: Learning from Data What is Machine Learning From Learning to Machine Learning learning: acquiring skill with experience accumulated from observations observations learning skill machine learning: acquiring skill with experience accumulated/computed from data data ML skill What is skill? Hsuan-Tien Lin (NTU CSIE) Quick Tour of Machine Learning 4/128 , machine learning: Learning from Data What is Machine Learning A More Concrete Definition skill improve some performance measure (e.g.
  • Understanding Support Vector Machines

    Understanding Support Vector Machines

    Chapter 7 Applying the same steps to compare the predicted values to the true values, we now obtain a correlation around 0.80, which is a considerable improvement over the previous result: > model_results2 <- compute(concrete_model2, concrete_test[1:8]) > predicted_strength2 <- model_results2$net.result > cor(predicted_strength2, concrete_test$strength) [,1] [1,] 0.801444583 Interestingly, in the original publication, I-Cheng Yeh reported a mean correlation of 0.885 using a very similar neural network. For some reason, we fell a bit short. In our defense, he is a civil engineering professor; therefore, he may have applied some subject matter expertise to the data preparation. If you'd like more practice with neural networks, you might try applying the principles learned earlier in this chapter to beat his result, perhaps by using different numbers of hidden nodes, applying different activation functions, and so on. The ?neuralnet help page provides more information on the various parameters that can be adjusted. Understanding Support Vector Machines A Support Vector Machine (SVM) can be imagined as a surface that defnes a boundary between various points of data which represent examples plotted in multidimensional space according to their feature values. The goal of an SVM is to create a fat boundary, called a hyperplane, which leads to fairly homogeneous partitions of data on either side. In this way, SVM learning combines aspects of both the instance-based nearest neighbor learning presented in Chapter 3, Lazy Learning – Classifcation Using Nearest Neighbors, and the linear regression modeling described in Chapter 6, Forecasting Numeric Data – Regression Methods. The combination is extremely powerful, allowing SVMs to model highly complex relationships.
  • Using a Support Vector Machine to Analyze a DNA Microarray∗

    Using a Support Vector Machine to Analyze a DNA Microarray∗

    Using a Support Vector Machine to Analyze a DNA Microarray∗ R. Morelli Department of Computer Science, Trinity College Hartford, CT 06106, USA [email protected] January 15, 2008 1 Introduction A DNA microarray is a small silicon chip that is covered with thousands of spots of DNA of known sequence. Biologists use microarrays to study gene expression patterns in a wide range of medical and scientific applications. Analysis of microarrays has led to discovery of genomic patterns that distinguish cancerous from non-cancerous cells. Other researchers have used microarrays to identify the genes in the brains of fish that are associated with certain kinds of mating behavior. Microarray data sets are very large and can only be analyzed with the help of computers. A support vector machine (SVM) is a powerful machine learning technique that is used in a variety of data mining applications, including the analysis of DNA microarrays. 2 Project Overview In this project we will learn how to use an SVM to recognize patterns in microarray data. Using an open source software package named libsvm and downloaded data from a cancer research project, we will to train an SVM to distinguish between gene expression patterns that are associated with two different forms of leukemia. Although this project will focus on SVMs and machine learning techniques, we will cover basic concepts in biology and genetics that underlie DNA analysis. We will gain experience using SVM software and emerge from this project with an improved understanding of how machine learning can be used to recognize important patterns in vast amounts of data.
  • Arxiv:1612.07993V1 [Stat.ML] 23 Dec 2016 Until Recently, No Package Providing Multiple Semi-Supervised Learning Meth- Ods Was Available in R

    Arxiv:1612.07993V1 [Stat.ML] 23 Dec 2016 Until Recently, No Package Providing Multiple Semi-Supervised Learning Meth- Ods Was Available in R

    RSSL: Semi-supervised Learning in R Jesse H. Krijthe1,2 1 Pattern Recognition Laboratory, Delft University of Technology 2 Department of Molecular Epidemiology, Leiden University Medical Center [email protected] Abstract. In this paper, we introduce a package for semi-supervised learning research in the R programming language called RSSL. We cover the purpose of the package, the methods it includes and comment on their use and implementation. We then show, using several code examples, how the package can be used to replicate well-known results from the semi-supervised learning literature. Keywords: Semi-supervised Learning, Reproducibility, Pattern Recog- nition, R Introduction Semi-supervised learning is concerned with using unlabeled examples, that is, examples for which we know the values for the input features but not the corre- sponding outcome, to improve the performance of supervised learning methods that only use labeled examples to train a model. An important motivation for investigations into these types of algorithms is that in some applications, gath- ering labels is relatively expensive or time-consuming, compared to the cost of obtaining an unlabeled example. Consider, for instance, building a web-page classifier. Downloading millions of unlabeled web-pages is easy. Reading them to assign a label is time-consuming. Effectively using unlabeled examples to improve supervised classifiers can therefore greatly reduce the cost of building a decently performing prediction model, or make it feasible in cases where labeling many examples is not a viable option. While the R programming language [] offers a rich set of implementations of a plethora of supervised learning methods, brought together by machine learning packages such as caret and mlr there are fewer implementations of methods that can deal with the semi-supervised learning setting.
  • Arxiv:1210.6293V1 [Cs.MS] 23 Oct 2012 So the Finally, Accessible

    Arxiv:1210.6293V1 [Cs.MS] 23 Oct 2012 So the Finally, Accessible

    Journal of Machine Learning Research 1 (2012) 1-4 Submitted 9/12; Published x/12 MLPACK: A Scalable C++ Machine Learning Library Ryan R. Curtin [email protected] James R. Cline [email protected] N. P. Slagle [email protected] William B. March [email protected] Parikshit Ram [email protected] Nishant A. Mehta [email protected] Alexander G. Gray [email protected] College of Computing Georgia Institute of Technology Atlanta, GA 30332 Editor: Abstract MLPACK is a state-of-the-art, scalable, multi-platform C++ machine learning library re- leased in late 2011 offering both a simple, consistent API accessible to novice users and high performance and flexibility to expert users by leveraging modern features of C++. ML- PACK provides cutting-edge algorithms whose benchmarks exhibit far better performance than other leading machine learning libraries. MLPACK version 1.0.3, licensed under the LGPL, is available at http://www.mlpack.org. 1. Introduction and Goals Though several machine learning libraries are freely available online, few, if any, offer efficient algorithms to the average user. For instance, the popular Weka toolkit (Hall et al., 2009) emphasizes ease of use but scales poorly; the distributed Apache Mahout library offers scal- ability at a cost of higher overhead (such as clusters and powerful servers often unavailable to the average user). Also, few libraries offer breadth; for instance, libsvm (Chang and Lin, 2011) and the Tilburg Memory-Based Learner (TiMBL) are highly scalable and accessible arXiv:1210.6293v1 [cs.MS] 23 Oct 2012 yet each offer only a single method.
  • Sb2b Statistical Machine Learning Hilary Term 2017

    Sb2b Statistical Machine Learning Hilary Term 2017

    SB2b Statistical Machine Learning Hilary Term 2017 Mihaela van der Schaar and Seth Flaxman Guest lecturer: Yee Whye Teh Department of Statistics Oxford Slides and other materials available at: http://www.oxford-man.ox.ac.uk/~mvanderschaar/home_ page/course_ml.html Administrative details Course Structure MMath Part B & MSc in Applied Statistics Lectures: Wednesdays 12:00-13:00, LG.01. Thursdays 16:00-17:00, LG.01. MSc: 4 problem sheets, discussed at the classes: weeks 2,4,6,7 (check website) Part C: 4 problem sheets Class Tutors: Lloyd Elliott, Kevin Sharp, and Hyunjik Kim Please sign up for the classes on the sign up sheet! Administrative details Course Aims 1 Understand statistical fundamentals of machine learning, with a focus on supervised learning (classification and regression) and empirical risk minimisation. 2 Understand difference between generative and discriminative learning frameworks. 3 Learn to identify and use appropriate methods and models for given data and task. 4 Learn to use the relevant R or python packages to analyse data, interpret results, and evaluate methods. Administrative details SyllabusI Part I: Introduction to supervised learning (4 lectures) Empirical risk minimization Bias/variance, Generalization, Overfitting, Cross validation Regularization Logistic regression Neural networks Part II: Classification and regression (3 lectures) Generative vs. Discriminative models K-nearest neighbours, Maximum Likelihood Estimation, Mixture models Naive Bayes, Decision trees, CART Support Vector Machines Random forest, Boostrap Aggregation (Bagging), Ensemble learning Expectation Maximization Administrative details SyllabusII Part III: Theoretical frameworks Statistical learning theory Decision theory Part IV: Further topics Optimisation Hidden Markov Models Backward-forward algorithms Reinforcement learning Overview Statistical Machine Learning What is Machine Learning? http://gureckislab.org Tom Mitchell, 1997 Any computer program that improves its performance at some task through experience.
  • Xgboost: a Scalable Tree Boosting System

    Xgboost: a Scalable Tree Boosting System

    XGBoost: A Scalable Tree Boosting System Tianqi Chen Carlos Guestrin University of Washington University of Washington [email protected] [email protected] ABSTRACT problems. Besides being used as a stand-alone predictor, it Tree boosting is a highly effective and widely used machine is also incorporated into real-world production pipelines for learning method. In this paper, we describe a scalable end- ad click through rate prediction [15]. Finally, it is the de- to-end tree boosting system called XGBoost, which is used facto choice of ensemble method and is used in challenges widely by data scientists to achieve state-of-the-art results such as the Netflix prize [3]. on many machine learning challenges. We propose a novel In this paper, we describe XGBoost, a scalable machine learning system for tree boosting. The system is available sparsity-aware algorithm for sparse data and weighted quan- 2 tile sketch for approximate tree learning. More importantly, as an open source package . The impact of the system has we provide insights on cache access patterns, data compres- been widely recognized in a number of machine learning and sion and sharding to build a scalable tree boosting system. data mining challenges. Take the challenges hosted by the machine learning competition site Kaggle for example. A- By combining these insights, XGBoost scales beyond billions 3 of examples using far fewer resources than existing systems. mong the 29 challenge winning solutions published at Kag- gle's blog during 2015, 17 solutions used XGBoost. Among these solutions, eight solely used XGBoost to train the mod- Keywords el, while most others combined XGBoost with neural net- Large-scale Machine Learning s in ensembles.
  • Computational Tools for Big Data — Python Libraries

    Computational Tools for Big Data — Python Libraries

    Computational Tools for Big Data | Python Libraries Finn Arup˚ Nielsen DTU Compute Technical University of Denmark September 15, 2015 Computational Tools for Big Data | Python libraries Overview Numpy | numerical arrays with fast computation Scipy | computation science functions Scikit-learn (sklearn) | machine learning Pandas | Annotated numpy arrays Cython | write C program in Python Finn Arup˚ Nielsen 1 September 15, 2015 Computational Tools for Big Data | Python libraries Python numerics The problem with Python: >>> [1, 2, 3] * 3 Finn Arup˚ Nielsen 2 September 15, 2015 Computational Tools for Big Data | Python libraries Python numerics The problem with Python: >>> [1, 2, 3] * 3 [1, 2, 3, 1, 2, 3, 1, 2, 3]# Not [3, 6, 9]! Finn Arup˚ Nielsen 3 September 15, 2015 Computational Tools for Big Data | Python libraries Python numerics The problem with Python: >>> [1, 2, 3] * 3 [1, 2, 3, 1, 2, 3, 1, 2, 3]# Not [3, 6, 9]! >>> [1, 2, 3] + 1# Wants [2, 3, 4]... Finn Arup˚ Nielsen 4 September 15, 2015 Computational Tools for Big Data | Python libraries Python numerics The problem with Python: >>> [1, 2, 3] * 3 [1, 2, 3, 1, 2, 3, 1, 2, 3]# Not [3, 6, 9]! >>> [1, 2, 3] + 1# Wants [2, 3, 4]... Traceback(most recent call last): File"<stdin>", line 1, in<module> TypeError: can only concatenate list(not"int") to list Finn Arup˚ Nielsen 5 September 15, 2015 Computational Tools for Big Data | Python libraries Python numerics The problem with Python: >>> [1, 2, 3] * 3 [1, 2, 3, 1, 2, 3, 1, 2, 3]# Not [3, 6, 9]! >>> [1, 2, 3] + 1# Wants [2, 3, 4]..
  • LIBSVM: a Library for Support Vector Machines

    LIBSVM: a Library for Support Vector Machines

    LIBSVM: A Library for Support Vector Machines Chih-Chung Chang and Chih-Jen Lin Department of Computer Science National Taiwan University, Taipei, Taiwan Email: [email protected] Initial version: 2001 Last updated: January 20, 2021 Abstract LIBSVM is a library for Support Vector Machines (SVMs). We have been actively developing this package since the year 2000. The goal is to help users to easily apply SVM to their applications. LIBSVM has gained wide popu- larity in machine learning and many other areas. In this article, we present all implementation details of LIBSVM. Issues such as solving SVM optimiza- tion problems, theoretical convergence, multi-class classification, probability estimates, and parameter selection are discussed in detail. Keywords: Classification, LIBSVM, optimization, regression, support vector ma- chines, SVM 1 Introduction Support Vector Machines (SVMs) are a popular machine learning method for classifi- cation, regression, and other learning tasks. Since the year 2000, we have been devel- oping the package LIBSVM as a library for support vector machines. The Web address of the package is at http://www.csie.ntu.edu.tw/~cjlin/libsvm. LIBSVM is cur- rently one of the most widely used SVM software. In this article,1 we present all implementation details of LIBSVM. However, this article does not intend to teach the practical use of LIBSVM. For instructions of using LIBSVM, see the README file included in the package, the LIBSVM FAQ,2 and the practical guide by Hsu et al. (2003). An earlier version of this article was published in Chang and Lin (2011). LIBSVM supports the following learning tasks.
  • Likelihood Algorithm Afterapplying PCA on LISS 3 Image Using Python Platform

    Likelihood Algorithm Afterapplying PCA on LISS 3 Image Using Python Platform

    International Journal of Scientific & Engineering Research, Volume 7, Issue 8, August-2016 1624 ISSN 2229-5518 Performance Evaluation of Maximum- Likelihood Algorithm afterApplying PCA on LISS 3 Image using Python Platform MayuriChatse, Vikas Gore, RasikaKalyane, Dr. K. V. Kale Dr. BabasahebAmbedkarMarathwada, University, Aurangabad, Maharashtra, India Abstract—This paper presents a system for multispectral image processing of LISS3 image using python programming language. The system starts with preprocessing technique i.e. by applying PCA on LISS3 image.PCA reduces image dimensionality by processing new, uncorrelated bands consisting of the principal components (PCs) of the given bands. Thus, this method transforms the original data set into a new dataset, which captures essential information.Further classification is done through maximum- likelihood algorithm. It is tested and evaluated on selected area i.e. region of interest. Accuracy is evaluated by confusion matrix and kappa statics.Satellite imagery tool using python shows better result in terms of overall accuracy and kappa coefficient than ENVI tool’s for maximum-likelihood algorithm. Keywords— Kappa Coefficient,Maximum-likelihood, Multispectral image,Overall Accuracy, PCA, Python, Satellite imagery tool —————————— —————————— I. Introduction microns), which provides information with a spatial resolution of 23.5 m not like in IRS-1C/1D (where the spatial resolution is 70.5 m) [12]. Remote sensing has become avitaltosol in wide areas. The classification of multi-spectral images has been a productive application that’s used for classification of land cover maps [1], urban growth [2], forest change [3], monitoring change in ecosystem condition [4], monitoring assessing deforestation and burned forest areas [5], agricultural expansion [6], The Data products are classified as Standard and have a mapping [7], real time fire detection [8], estimating tornado system level accuracy.
  • Scikit-Learn

    Scikit-Learn

    Scikit-Learn i Scikit-Learn About the Tutorial Scikit-learn (Sklearn) is the most useful and robust library for machine learning in Python. It provides a selection of efficient tools for machine learning and statistical modeling including classification, regression, clustering and dimensionality reduction via a consistence interface in Python. This library, which is largely written in Python, is built upon NumPy, SciPy and Matplotlib. Audience This tutorial will be useful for graduates, postgraduates, and research students who either have an interest in this Machine Learning subject or have this subject as a part of their curriculum. The reader can be a beginner or an advanced learner. Prerequisites The reader must have basic knowledge about Machine Learning. He/she should also be aware about Python, NumPy, Scipy, Matplotlib. If you are new to any of these concepts, we recommend you take up tutorials concerning these topics, before you dig further into this tutorial. Copyright & Disclaimer Copyright 2019 by Tutorials Point (I) Pvt. Ltd. All the content and graphics published in this e-book are the property of Tutorials Point (I) Pvt. Ltd. The user of this e-book is prohibited to reuse, retain, copy, distribute or republish any contents or a part of contents of this e-book in any manner without written consent of the publisher. We strive to update the contents of our website and tutorials as timely and as precisely as possible, however, the contents may contain inaccuracies or errors. Tutorials Point (I) Pvt. Ltd. provides no guarantee regarding the accuracy, timeliness or completeness of our website or its contents including this tutorial.
  • Liblinear: a Library for Large Linear Classification

    Liblinear: a Library for Large Linear Classification

    Journal of Machine Learning Research 9 (2008) 1871-1874 Submitted 5/08; Published 8/08 LIBLINEAR: A Library for Large Linear Classification Rong-En Fan [email protected] Kai-Wei Chang [email protected] Cho-Jui Hsieh [email protected] Xiang-Rui Wang [email protected] Chih-Jen Lin [email protected] Department of Computer Science National Taiwan University Taipei 106, Taiwan Editor: Soeren Sonnenburg Abstract LIBLINEAR is an open source library for large-scale linear classification. It supports logistic regres- sion and linear support vector machines. We provide easy-to-use command-line tools and library calls for users and developers. Comprehensive documents are available for both beginners and advanced users. Experiments demonstrate that LIBLINEAR is very efficient on large sparse data sets. Keywords: large-scale linear classification, logistic regression, support vector machines, open source, machine learning 1. Introduction Solving large-scale classification problems is crucial in many applications such as text classification. Linear classification has become one of the most promising learning techniques for large sparse data with a huge number of instances and features. We develop LIBLINEAR as an easy-to-use tool to deal with such data. It supports L2-regularized logistic regression (LR), L2-loss and L1-loss linear support vector machines (SVMs) (Boser et al., 1992). It inherits many features of the popular SVM library LIBSVM (Chang and Lin, 2001) such as simple usage, rich documentation, and open source license (the BSD license1). LIBLINEAR is very efficient for training large-scale problems. For example, it takes only several seconds to train a text classification problem from the Reuters Corpus Volume 1 (rcv1) that has more than 600,000 examples.