Summary: ROCR is a package for evaluating and visualizing the performance of scoring classifiers in the statistical language R. It features over 25 performance measures that can be freely combined to create two-dimensional performance curves. Standard methods for investigating trade-offs between specific performance measures are available within a uniform framework, including receiver operating characteristic (ROC) graphs, precision/recall plots, lift charts and cost curves. ROCR integrates tightly with R's powerful graphics capabilities, thus allowing for highly adjustable plots. Being equipped with only three commands and reasonable default values for optional parameters, ROCR combines flexibility with ease of usage.

Availability:http://rocr.bioinf.mpi-sb.mpg.de. ROCR can be used under the terms of the GNU General Public License. Running within R, it is platform-independent.

Contact: tobias.sing@mpi-sb.mpg.de

ROCR: visualizing classifier performance in R

/pdf/editorial-special-issue-on-learning-from-imbalanced-data-332aig47dl.pdf

Editorial: special issue on learning from imbalanced data sets

Undersampling is a popular method in dealing with class-imbalance problems, which uses only a subset of the majority class and thus is very efficient. The main deficiency is that many majority class examples are ignored. We propose two algorithms to overcome this deficiency. EasyEnsemble samples several subsets from the majority class, trains a learner using each of them, and combines the outputs of those learners. BalanceCascade trains the learners sequentially, where in each step, the majority class examples that are correctly classified by the current trained learners are removed from further consideration. Experimental results show that both methods have higher Area Under the ROC Curve, F-measure, and G-mean values than many existing class-imbalance learning methods. Moreover, they have approximately the same training time as that of undersampling when the same number of weak classifiers is used, which is significantly faster than other methods.

Exploratory Undersampling for Class-Imbalance Learning

Convolutional Neural Networks (CNNs) have proven very effective in image classification and show promise for audio. We use various CNN architectures to classify the soundtracks of a dataset of 70M training videos (5.24 million hours) with 30,871 video-level labels. We examine fully connected Deep Neural Networks (DNNs), AlexNet [1], VGG [2], Inception [3], and ResNet [4]. We investigate varying the size of both training set and label vocabulary, finding that analogs of the CNNs used in image classification do well on our audio classification task, and larger training and label sets help up to a point. A model using embeddings from these classifiers does much better than raw features on the Audio Set [5] Acoustic Event Detection (AED) classification task.

CNN architectures for large-scale audio classification

With the increasing advances in hardware technology for data collection, and advances in software technology (databases) for data organization, computer scientists have increasingly participated in the latest advancements of the outlier analysis field. Computer scientists, specifically, approach this field based on their practical experiences in managing large amounts of data, and with far fewer assumptions the data can be of any type, structured or unstructured, and may be extremely large. Outlier Analysisis a comprehensive exposition, as understood by data mining experts, statisticians and computer scientists. The book has been organized carefully, and emphasis was placed on simplifying the content, so that students and practitioners can also benefit. Chapters will typically cover one of three areas: methods and techniques commonly used in outlier analysis, such as linear methods, proximity-based methods, subspace methods, and supervised methods; data domains, such as, text, categorical, mixed-attribute, time-series, streaming, discrete sequence, spatial and network data; and key applications of these methods as applied to diverse domains such as credit card fraud detection, intrusion detection, medical diagnosis, earth science, web log analytics, and social network analysis are covered.

Outlier Analysis

Receiver Operating Characteristics (ROC) graphs are a useful technique for organizing classifiers and visualizing their performance. ROC graphs are commonly used in medical decision making, and in recent years have been increasingly adopted in the machine learning and data mining research communities. Although ROC graphs are apparently simple, there are some common misconceptions and pitfalls when using them in practice. This article serves both as a tutorial introduction to ROC graphs and as a practical guide for using them in research.

https://www.eecs.wsu.edu/~holder/courses/CptS570/fall07/papers/Fawcett04.pdf

What's In A Name? Malay Seals As Onomastic Sources

With their spendid illumination, elegant calligraphy, intricate seals and courtly and refined language, the Malay letters embody all that is most beautiful in Malay culture, civilization and aesthetics. Indeed, such letters offer a glimpse of the complex social and political structure of Malay society, for every element - from the position of the heading and seal to the choice of yellow silk envelope - was determined by the relative rank of the correspondents and the diplomatic importance of the missive. Displayed and translated in this exhibition catalogue are 100 Malay letters spanning nearly four centuries, from the oldest known Malay manuscript of 1521, through to the eve of the 20th century. The objects are not only beautiful art objects but also a resource for the political, diplomatic and economic history of the Malay archipelago. Also presented here are traditional Malay manuscript guides to letter-writing, and literary and historical texts which describe the elaborate ceremonies which accompanied the sending of royal Malay letters, and the status and quality of the envoys entrusted to carry and deliver them. The Malay letters form a legacy whose true value has come to be recognized only slowly, but which is all the more stunning for its discovery.

The legacy of the Malay letter = Warisan warkah Melayu

This special issue of Indonesia and the Malay World was compiled by friends and colleagues as a tribute to Professor E. Ulrich Kratz's three decades of teaching Jawi and traditional Malay literature at the School of Oriental and African Studies, University of London, and to mark his 70th birthday on 14 October 2014. Reflecting Ulrich's deep interest in Malay manuscript texts and letters over many years (see the list of publications compiled by Helen Cordell in this issue), this Festschrift takes the rather unusual form of a compilation of reproductions of Malay manuscripts in Jawi script, accompanied by commentaries on the handwriting and spelling. Nearly all the manuscripts are dated or firmly dateable, and come from known locations. The hope is that this issue will be useful as a sourcebook for the study of the development of Jawi script, and in particular its palaeography and orthography, over the course of nearly three and a half centuries. The manuscripts presented date from the end of the 16th centu...

Annabel Teh Gallop

Papers

What's In A Name? Malay Seals As Onomastic Sources

The legacy of the Malay letter = Warisan warkah Melayu

A jawi sourcebook for the study of malay palaeography and orthography

Golden Letters: Writing Traditions of Indonesia

Golden Letters: Writing Traditions of Indonesia / Surat Emas: Budaya Tulis di Indonesia