Skip to main content

    Mark Hall

    University of Waikato, Department of Computer Science, Adjunct
    Papers
    Flow Clustering Using Machine Learning Techniquesmore
    by James Brunskill and Mark Hall
    Packet header traces are widely used in network analysis. Header traces are the aggregate of traffic from many concurrent applications. We present a methodology, based on machine learning, that can break the trace down into clusters of... more
    Packet header traces are widely used in network analysis. Header traces are the aggregate of traffic from many concurrent applications. We present a methodology, based on machine learning, that can break the trace down into clusters of traffic where each cluster has different traffic characteristics. Typical clusters include bulk transfer, single and multiple transactions and interactive traffic, amongst others. The paper includes a description of the methodology, a visualisation of the attribute statistics that aids in recognising cluster types and a discussion of the stability and effectiveness of the methodology.
    Publication Date: 2004
    Publication Name: Passive and Active Network Measurement
    Research Interests:
    Download (.pdf)
    Data mining: practical machine learning tools and techniques with Java implementationsmore
    by I. Witten and Mark Hall
    Data Mining: Practical Machine Learning Tools and ... Witten and Frank's textbook was one of two books that 1 used for a data mining class in the Fall of 2001. ...
    Publisher: portal.acm.org
    Publication Name: ACM SIGMOD Record
    Research Interests:
    Gene selection from microarray data for cancer classification—a machine learning approachmore
    by Igor Tetko and Mark Hall
    Publication Date: 2005
    Publication Name: Computational Biology and Chemistry
    Research Interests:
    Download (.pdf)
    The WEKA Data Mining Software: An Updatemore
    by Peter Reutemann, Ian Witten, Mark Hall, Mark Hall, and Eibe Frank
    More than twelve years have elapsed since the first public release of WEKA. In that time, the software has been re- written entirely from scratch, evolved substantially and now accompanies a text on data mining (35). These days, WEKA... more
    More than twelve years have elapsed since the first public release of WEKA. In that time, the software has been re- written entirely from scratch, evolved substantially and now accompanies a text on data mining (35). These days, WEKA enjoys widespread acceptance in both academia and busi- ness, has an active community, and has been downloaded more than 1.4 million
    Publication Date: 2000
    Research Interests:
    Download (.pdf)
    Using Classification to Evaluate the Output of Confidence-Based Association Rule Miningmore
    by Mark Hall and Eibe Frank
    Association rule mining is a data mining technique that reveals interesting relationships in a database. Existing approaches employ different parameters to search for interesting rules. This fact and the large number of rules make it... more
    Association rule mining is a data mining technique that reveals interesting relationships in a database. Existing approaches employ different parameters to search for interesting rules. This fact and the large number of rules make it difficult to compare the output of confidence ...
    Publication Date: 2004
    Publication Name: Lecture Notes in Computer Science
    Research Interests:
    Weka-A Machine Learning Workbench for Data Miningmore
    by Ian Witten, Eibe Frank, and Mark Hall
    Publication Date: 2009
    Publication Name: Data Mining and Knowledge Discovery Handbook
    Research Interests:
    Download (.pdf)
    Visualizing Class Probability Estimatorsmore
    by Mark Hall and Eibe Frank
    ABSTRACT Inducing classifiers that make accurate predictions on future data is a driving force for research in inductive learning. However, also of importance to the users is how to gain information from the models produced.... more
    ABSTRACT Inducing classifiers that make accurate predictions on future data is a driving force for research in inductive learning. However, also of importance to the users is how to gain information from the models produced. Unfortunately, some of the most powerful inductive learning algorithms generate “black boxes”—that is, the representation of the model makes it virtually impossible to gain any insight into what has been learned. This paper presents a technique that can help the user understand why a classifier makes the predictions that it does by providing a two-dimensional visualization of its class probability estimates. It requires the classifier to generate class probabilities but most practical algorithms are able to do so (or can be modified to this end).
    Publication Date: 2003
    Publication Name: Lecture Notes in Computer Science
    A Simple Approach to Ordinal Classificationmore
    by Mark Hall and Eibe Frank
    Publication Date: 2001
    Publication Name: Lecture Notes in Computer Science
    Research Interests:
    Download (.pdf)
    Additive Regression Applied to a Large-Scale Collaborative Filtering Problemmore
    by Mark Hall and Eibe Frank
    ABSTRACT The much-publicized Netflix competition has put the spotlight on the application domain of collaborative filtering and has sparked interest in machine learning algorithms that can be applied to this sort of problem. The demanding... more
    ABSTRACT The much-publicized Netflix competition has put the spotlight on the application domain of collaborative filtering and has sparked interest in machine learning algorithms that can be applied to this sort of problem. The demanding nature of the Netflix data has lead to some interesting and ingenious modifications to standard learning methods in the name of efficiency and speed. There are three basic methods that have been applied in most approaches to the Netflix problem so far: stand-alone neighborhood-based methods, latent factor models based on singular-value decomposition, and ensembles consisting of variations of these techniques. In this paper we investigate the application of forward stage-wise additive modeling to the Netflix problem, using two regression schemes as base learners: ensembles of weighted simple linear regressors and k-means clustering—the latter being interpreted as a tool for multi-variate regression in this context. Experimental results show that our methods produce competitive results.
    Publication Date: 2008
    Publication Name: Lecture Notes in Computer Science
    Racing Committees for Large Datasetsmore
    by Mark Hall and Eibe Frank
    Publication Date: 2002
    Publication Name: Lecture Notes in Computer Science
    Research Interests:
    Download (.pdf)
    Multiclass Alternating Decision Treesmore
    by Mark Hall and Eibe Frank
    Publication Date: 2002
    Publication Name: Lecture Notes in Computer Science
    Research Interests:
    Download (.pdf)
    Large-scale attribute selection using wrappersmore
    by Mark Hall and Eibe Frank
    Publication Date: 2009
    Publication Name: 2009 IEEE Symposium on Computational Intelligence and Data Mining
    Research Interests:
    Download (.pdf)
    Data Mining: What's iIt all about?; 2. Input: concepts, instances, and attributes; 3. Output: knowledge representation; 4. Algorithms: the basic methods; 5. Credibility: evaluating what's been learned. Part II. Advanced Data Mining: 6. Implementations: real machine learning schemes; 7. Data trans...more
    by Mark Hall
    Experiences with a weighted decision tree learnermore
    by Mark Hall
    ABSTRACT
    WEKA KnowledgeFlow Tutorial for Version 3-5-8more
    by Peter Reutemann and Mark Hall
    WEKA Manual for Version 3-7-8more
    by Mark Hall and Peter Reutemann
    Speeding Up Logistic Model Tree Inductionmore
    by Mark Hall and Eibe Frank
    Logistic Model Trees have been shown to be very accurate and compact classifiers [8]. Their greatest disadvantage is the computational complexity of inducing the logistic regression models in the tree. We address this issue by using the... more
    Logistic Model Trees have been shown to be very accurate and compact classifiers [8]. Their greatest disadvantage is the computational complexity of inducing the logistic regression models in the tree. We address this issue by using the AIC criterion [1] instead of cross-...
    Publication Date: 2005
    Publication Name: Lecture Notes in Computer Science
    The WEKA data mining softwaremore
    by Ian Witten, Eibe Frank, Mark Hall, and Peter Reutemann
    Research Interests:
    Download (.pdf)
    The positive effects of negative information: Extending one-class classification models in binary proteomic sequence classificationmore
    by Mark Hall
    ABSTRACT={Profile Hidden Markov Models (PHMMs) have been widely used as models for Multiple Sequence Alignments. By their nature, they are generative one-class classifiers trained only on sequences belonging to the target class they... more
    ABSTRACT={Profile Hidden Markov Models (PHMMs) have been widely used as models for Multiple Sequence Alignments. By their nature, they are generative one-class classifiers trained only on sequences belonging to the target class they represent. Nevertheless, they ...
    Publication Date: 2009
    Publication Name: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Generating Rule Sets from Model Treesmore
    by Mark Hall and Eibe Frank
    Publication Date: 1999
    Publication Name: Lecture Notes in Computer Science
    Research Interests:
    Download (.pdf)
    Combining Naive Bayes and Decision Tablesmore
    by Mark Hall
    We investigate a simple semi-naive Bayesian ranking method that combines naive Bayes with induction of decision tables. Naive Bayes and decision tables can both be trained effi- ciently, and the same holds true for the combined semi-naive... more
    We investigate a simple semi-naive Bayesian ranking method that combines naive Bayes with induction of decision tables. Naive Bayes and decision tables can both be trained effi- ciently, and the same holds true for the combined semi-naive model. We show that the resulting ranker, compared to ei- ther component technique, frequently significantly increases AUC. For some datasets it significantly improves on both techniques. This is also the case when attribute selection is performed in naive Bayes and its semi-naive variant.
    Research Interests:
    Download (.pdf)
    Logistic Model Treesmore
    by Mark Hall and Eibe Frank
    Publication Date: 2005
    Publication Name: Machine Learning
    Research Interests:
    Download ()
    Comparison of consumer and producer perceptions of mushroom qualitymore
    by Mark Hall
    Publication Date: 2002
    A development environment for predictive modelling in foodsmore
    by Mark Hall
    Publication Date: 2002
    Publication Name: International Journal of Food Microbiology
    Research Interests:
    Download ()
    Benchmarking attribute selection techniques for discrete class data miningmore
    by Mark Hall
    Publication Date: 2003
    Publication Name: IEEE Transactions on Knowledge and Data Engineering
    Research Interests:
    Download (.pdf)
    Gene selection from microarray data for cancer classification—a machine learning approachmore
    by Mark Hall and Eibe Frank
    Publication Date: 2005
    Publication Name: Computational Biology and Chemistry
    Research Interests:
    Download (.pdf)
    Data mining in bioinformatics using Wekamore
    by Ian Witten, Eibe Frank, and Mark Hall
    Publication Date: 2004
    Publication Name: Bioinformatics
    Research Interests:
    Download (.pdf)
    Wekamore
    by Mark Hall and Eibe Frank
    The Weka workbench is an organized collection of state-of-the-art machine lear-ning algorithms and data preprocessing tools. The basic way of interacting with these methods is by invoking them from the command line. However, convenient... more
    The Weka workbench is an organized collection of state-of-the-art machine lear-ning algorithms and data preprocessing tools. The basic way of interacting with these methods is by invoking them from the command line. However, convenient interactive graphical user ...
    Publisher: Springer
    Publication Date: 2005
    Publication Name: Data Mining and …
    Research Interests:
    Investigation of association models to describe consumer purchase patternsmore
    by Mark Hall
    Publication Date: 2001
    A logic boosting approach to inducing multiclass alternating decision treesmore
    by Mark Hall
    Publication Date: 2002
    Data mining and knowledge discovery handbookmore
    by Mark Hall
    ... Vazirgiannis 1. Data Pre-processing and Quality Assessment 663 2. Evaluation of Classification Methods 664 3. Association Rules 671 4. Cluster Validity 675 References 694 31 Data Mining Model Comparison 697 Paolo Giudici 1. Data... more
    ... Vazirgiannis 1. Data Pre-processing and Quality Assessment 663 2. Evaluation of Classification Methods 664 3. Association Rules 671 4. Cluster Validity 675 References 694 31 Data Mining Model Comparison 697 Paolo Giudici 1. Data Mining and Statistics 697 2. Data Mining ...
    Publication Date: 2010
    Research Interests:
    Determining progression in glaucoma using visual fieldsmore
    by Mark Hall and Eibe Frank
    Publication Date: 2001
    Research Interests:
    Download (.pdf)
    Locally weighted naive Bayesmore
    by Mark Hall and Eibe Frank
    Publication Date: 2003
    Research Interests:
    Download (.pdf)
    Data mining: practical machine learning tools and techniques with Java implementationsmore
    by Ian Witten, Eibe Frank, and Mark Hall
    Data Mining: Practical Machine Learning Tools and ... Witten and Frank's textbook was one of two books that 1 used for a data mining class in the Fall of 2001. ...
    Publisher: portal.acm.org
    Publication Name: ACM SIGMOD Record
    Research Interests:
    Download (.pdf)
    Interactive machine learning: letting users build classifiersmore
    by Ian Witten, Eibe Frank, and Mark Hall
    Publication Date: 2001
    Publication Name: International Journal of Human-Computer Studies
    Research Interests:
    Download (.pdf)
    -
    by 30-day views
    -
    total views
    -
    followers
    Related Authors
    Gualtiero Piccinini
    David Seamon
    Kyle Whyte
    Tiffany Jones
    Edy Chandra
    Pirjo Kristiina Virtanen
    Dr. Shadab Alam
    Diego Villar
    Miguel Aparicio
    So Miyagawa