Combining heterogeneous classifiers for relational databases

Practical usage of machine learning is gaining strategic importance in enterprises looking for business intelligence. However, most enterprise data is distributed in multiple relational databases with expert-designed schema. Using traditional single-table machine learning techniques over such data not only incur a computational penalty for converting to a flat form (mega-join), even the human-specified semantic information present in the relations is lost. In this paper, we present a practical, two-phase hierarchical meta-classification algorithm for relational databases with a semantic divide and conquer approach. We propose a recursive, prediction aggregation technique over heterogeneous classifiers applied on individual database tables. The proposed algorithm was evaluated on three diverse datasets. namely TPCH, PKDD and UCI benchmarks and showed considerable reduction in classification time without any loss of prediction accuracy. (C) 2012 Elsevier Ltd. All rights reserved.

Efficient Algorithms for Linear Summed Error Structural SVMs

Structural Support Vector Machines (SSVMs) have become a popular tool in machine learning for predicting structured objects like parse trees, Part-of-Speech (POS) label sequences and image segments. Various efficient algorithmic techniques have been proposed for training SSVMs for large datasets. The typical SSVM formulation contains a regularizer term and a composite loss term. The loss term is usually composed of the Linear Maximum Error (LME) associated with the training examples. Other alternatives for the loss term are yet to be explored for SSVMs. We formulate a new SSVM with Linear Summed Error (LSE) loss term and propose efficient algorithms to train the new SSVM formulation using primal cutting-plane method and sequential dual coordinate descent method. Numerical experiments on benchmark datasets demonstrate that the sequential dual coordinate descent method is faster than the cutting-plane method and reaches the steady-state generalization performance faster. It is thus a useful alternative for training SSVMs when linear summed error is used.

Efficient Methods for Robust Classification Under Uncertainty in Kernel Matrices

In this paper we study the problem of designing SVM classifiers when the kernel matrix, K, is affected by uncertainty. Specifically K is modeled as a positive affine combination of given positive semi definite kernels, with the coefficients ranging in a norm-bounded uncertainty set. We treat the problem using the Robust Optimization methodology. This reduces the uncertain SVM problem into a deterministic conic quadratic problem which can be solved in principle by a polynomial time Interior Point (IP) algorithm. However, for large-scale classification problems, IP methods become intractable and one has to resort to first-order gradient type methods. The strategy we use here is to reformulate the robust counterpart of the uncertain SVM problem as a saddle point problem and employ a special gradient scheme which works directly on the convex-concave saddle function. The algorithm is a simplified version of a general scheme due to Juditski and Nemirovski (2011). It achieves an O(1/T-2) reduction of the initial error after T iterations. A comprehensive empirical study on both synthetic data and real-world protein structure data sets show that the proposed formulations achieve the desired robustness, and the saddle point based algorithm outperforms the IP method significantly.

Infinite push: a new support vector ranking algorithm that directly optimizes accuracy at the absolute top of the list

Ranking problems have become increasingly important in machine learning and data mining in recent years, with applications ranging from information retrieval and recommender systems to computational biology and drug discovery. In this paper, we describe a new ranking algorithm that directly maximizes the number of relevant objects retrieved at the absolute top of the list. The algorithm is a support vector style algorithm, but due to the different objective, it no longer leads to a quadratic programming problem. Instead, the dual optimization problem involves l1, ∞ constraints; we solve this dual problem using the recent l1, ∞ projection method of Quattoni et al (2009). Our algorithm can be viewed as an l∞-norm extreme of the lp-norm based algorithm of Rudin (2009) (albeit in a support vector setting rather than a boosting setting); thus we refer to the algorithm as the ‘Infinite Push’. Experiments on real-world data sets confirm the algorithm’s focus on accuracy at the absolute top of the list.

Identification of coherent synchronous generators in a multi- machine power system using support vector clustering

This paper illustrates the application of a new technique, based on Support Vector Clustering (SVC) for the direct identification of coherent synchronous generators in a large interconnected Multi-Machine Power Systems. The clustering is based on coherency measures, obtained from the time domain responses of the generators following system disturbances. The proposed clustering algorithm could be integrated into a wide-area measurement system that enables fast identification of coherent clusters of generators for the construction of dynamic equivalent models. An application of the proposed method is demonstrated on a practical 15 generators 72-bus system, an equivalent of Indian Southern grid in an attempt to show the effectiveness of this clustering approach. The effects of short circuit fault locations on coherency are also investigated.

Support vector machine based high speed protection relay for EHV/UHV transmission line

This paper presents a fast and accurate relaying technique for a long 765kv UHV transmission line based on support vector machine. For a long EHV/UHV transmission line with large distributed capacitance, a traditional distance relay which uses a lumped parameter model of the transmission line can cause malfunction of the relay. With a frequency of 1kHz, 1/4th cycle of instantaneous values of currents and voltages of all phases at the relying end are fed to Support Vector Machine(SVM). The SVM detects fault type accurately using 3 milliseconds of post-fault data and reduces the fault clearing time which improves the system stability and power transfer capability. The performance of relaying scheme has been checked with a typical 765kV Indian transmission System which is simulated using the Electromagnetic Transients Program(EMTP) developed by authors in which the distributed parameter line model is used. More than 15,000 different short circuit fault cases are simulated by varying fault location, fault impedance, fault incidence angle and fault type to train the SVM for high speed accurate relaying. Simulation studies have shown that the proposed relay provides fast and accurate protection irrespective of fault location, fault impedance, incidence time of fault and fault type. And also the proposed scheme can be used as augmentation for the existing relaying, particularly for Zone-2, Zone-3 protection.

On the relation between K-means and PLSA

Non-negative matrix factorization [5](NMF) is a well known tool for unsupervised machine learning. It can be viewed as a generalization of the K-means clustering, Expectation Maximization based clustering and aspect modeling by Probabilistic Latent Semantic Analysis (PLSA). Specifically PLSA is related to NMF with KL-divergence objective function. Further it is shown that K-means clustering is a special case of NMF with matrix L2 norm based error function. In this paper our objective is to analyze the relation between K-means clustering and PLSA by examining the KL-divergence function and matrix L2 norm based error function.

Efficient classification using phrases generated by topic models

There are many popular models available for classification of documents like Naïve Bayes Classifier, k-Nearest Neighbors and Support Vector Machine. In all these cases, the representation is based on the “Bag of words” model. This model doesn't capture the actual semantic meaning of a word in a particular document. Semantics are better captured by proximity of words and their occurrence in the document. We propose a new “Bag of Phrases” model to capture this discriminative power of phrases for text classification. We present a novel algorithm to extract phrases from the corpus using the well known topic model, Latent Dirichlet Allocation(LDA), and to integrate them in vector space model for classification. Experiments show a better performance of classifiers with the new Bag of Phrases model against related representation models.

Identification of fault location in power distribution system with distributed generation using support vector machines

This paper presents a multi-class support vector machine (SVMs) approach for locating and diagnosing faults in electric power distribution feeders with the penetration of Distributed Generations (DGs). The proposed approach is based on the three phase voltage and current measurements which are available at all the sources i.e. substation and at the connection points of DG. To illustrate the proposed methodology, a practical distribution feeder emanating from 132/11kV-grid substation in India with loads and suitable number of DGs at different locations is considered. To show the effectiveness of the proposed methodology, practical situations in distribution systems (DS) such as all types of faults with a wide range of varying fault locations, source short circuit (SSC) levels and fault impedances are considered for studies. The proposed fault location scheme is capable of accurately identify the fault type, location of faulted feeder section and the fault impedance. The results demonstrate the feasibility of applying the proposed method in practical in smart grid distribution automation (DA) for fault diagnosis.

Differentially private stochastic gradient descent algorithm for multiparty classification

We consider the problem of developing privacy-preserving machine learning algorithms in a dis-tributed multiparty setting. Here different parties own different parts of a data set, and the goal is to learn a classifier from the entire data set with-out any party revealing any information about the individual data points it owns. Pathak et al [7]recently proposed a solution to this problem in which each party learns a local classifier from its own data, and a third party then aggregates these classifiers in a privacy-preserving manner using a cryptographic scheme. The generaliza-tion performance of their algorithm is sensitive to the number of parties and the relative frac-tions of data owned by the different parties. In this paper, we describe a new differentially pri-vate algorithm for the multiparty setting that uses a stochastic gradient descent based procedure to directly optimize the overall multiparty ob-jective rather than combining classifiers learned from optimizing local objectives. The algorithm achieves a slightly weaker form of differential privacy than that of [7], but provides improved generalization guarantees that do not depend on the number of parties or the relative sizes of the individual data sets. Experimental results corrob-orate our theoretical findings.

Multi-task learning using shared and task specific information

Multi-task learning solves multiple related learning problems simultaneously by sharing some common structure for improved generalization performance of each task. We propose a novel approach to multi-task learning which captures task similarity through a shared basis vector set. The variability across tasks is captured through task specific basis vector set. We use sparse support vector machine (SVM) algorithm to select the basis vector sets for the tasks. The approach results in a sparse model where the prediction is done using very few examples. The effectiveness of our approach is demonstrated through experiments on synthetic and real multi-task datasets.

Cluster labeling for multilingual scatter/gather using comparable corpora

Scatter/Gather systems are increasingly becoming useful in browsing document corpora. Usability of the present-day systems are restricted to monolingual corpora, and their methods for clustering and labeling do not easily extend to the multilingual setting, especially in the absence of dictionaries/machine translation. In this paper, we study the cluster labeling problem for multilingual corpora in the absence of machine translation, but using comparable corpora. Using a variational approach, we show that multilingual topic models can effectively handle the cluster labeling problem, which in turn allows us to design a novel Scatter/Gather system ShoBha. Experimental results on three datasets, namely the Canadian Hansards corpus, the entire overlapping Wikipedia of English, Hindi and Bengali articles, and a trilingual news corpus containing 41,000 articles, confirm the utility of the proposed system.

The Lovasz θ function, SVMs and finding large dense subgraphs

The Lovasz θ function of a graph, is a fundamental tool in combinatorial optimization and approximation algorithms. Computing θ involves solving a SDP and is extremely expensive even for moderately sized graphs. In this paper we establish that the Lovasz θ function is equivalent to a kernel learning problem related to one class SVM. This interesting connection opens up many opportunities bridging graph theoretic algorithms and machine learning. We show that there exist graphs, which we call SVM−θ graphs, on which the Lovasz θ function can be approximated well by a one-class SVM. This leads to a novel use of SVM techniques to solve algorithmic problems in large graphs e.g. identifying a planted clique of size Θ(n√) in a random graph G(n,12). A classic approach for this problem involves computing the θ function, however it is not scalable due to SDP computation. We show that the random graph with a planted clique is an example of SVM−θ graph, and as a consequence a SVM based approach easily identifies the clique in large graphs and is competitive with the state-of-the-art. Further, we introduce the notion of a ''common orthogonal labeling'' which extends the notion of a ''orthogonal labelling of a single graph (used in defining the θ function) to multiple graphs. The problem of finding the optimal common orthogonal labelling is cast as a Multiple Kernel Learning problem and is used to identify a large common dense region in multiple graphs. The proposed algorithm achieves an order of magnitude scalability compared to the state of the art.

Modified particle swarm optimization for pattern clustering

Data clustering is a common technique for statistical data analysis, which is used in many fields, including machine learning and data mining. Clustering is grouping of a data set or more precisely, the partitioning of a data set into subsets (clusters), so that the data in each subset (ideally) share some common trait according to some defined distance measure. In this paper we present the genetically improved version of particle swarm optimization algorithm which is a population based heuristic search technique derived from the analysis of the particle swarm intelligence and the concepts of genetic algorithms (GA). The algorithm combines the concepts of PSO such as velocity and position update rules together with the concepts of GA such as selection, crossover and mutation. The performance of the above proposed algorithm is evaluated using some benchmark datasets from Machine Learning Repository. The performance of our method is better than k-means and PSO algorithm.

Support vector clustering-based direct coherency identification of generators in a multi-machine power system

This study investigates the application of support vector clustering (SVC) for the direct identification of coherent synchronous generators in large interconnected multi-machine power systems. The clustering is based on coherency measure, which indicates the degree of coherency between any pair of generators. The proposed SVC algorithm processes the coherency measure matrix that is formulated using the generator rotor measurements to cluster the coherent generators. The proposed approach is demonstrated on IEEE 10 generator 39-bus system and an equivalent 35 generators, 246-bus system of practical Indian southern grid. The effect of number of data samples and fault locations are also examined for determining the accuracy of the proposed approach. An extended comparison with other clustering techniques is also included, to show the effectiveness of the proposed approach in grouping the data into coherent groups of generators. This effectiveness of the coherent clusters obtained with the proposed approach is compared in terms of a set of clustering validity indicators and in terms of statistical assessment that is based on the coherency degree of a generator pair.