SPSA Based Feature Relevance Estimation For Video Retrieval

With the availability of a huge amount of video data on various sources, efficient video retrieval tools are increasingly in demand. Video being a multi-modal data, the perceptions of ``relevance'' between the user provided query video (in case of Query-By-Example type of video search) and retrieved video clips are subjective in nature. We present an efficient video retrieval method that takes user's feedback on the relevance of retrieved videos and iteratively reformulates the input query feature vectors (QFV) for improved video retrieval. The QFV reformulation is done by a simple, but powerful feature weight optimization method based on Simultaneous Perturbation Stochastic Approximation (SPSA) technique. A video retrieval system with video indexing, searching and relevance feedback (RF) phases is built for demonstrating the performance of the proposed method. The query and database videos are indexed using the conventional video features like color, texture, etc. However, we use the comprehensive and novel methods of feature representations, and a spatio-temporal distance measure to retrieve the top M videos that are similar to the query. In feedback phase, the user activated iterative on the previously retrieved videos is used to reformulate the QFV weights (measure of importance) that reflect the user's preference, automatically. It is our observation that a few iterations of such feedback are generally sufficient for retrieving the desired video clips. The novel application of SPSA based RF for user-oriented feature weights optimization makes the proposed method to be distinct from the existing ones. The experimental results show that the proposed RF based video retrieval exhibit good performance.

Scalable multi-dimensional user intent identification using tree structured distributions

The problem of identifying user intent has received considerable attention in recent years, particularly in the context of improving the search experience via query contextualization. Intent can be characterized by multiple dimensions, which are often not observed from query words alone. Accurate identification of Intent from query words remains a challenging problem primarily because it is extremely difficult to discover these dimensions. The problem is often significantly compounded due to lack of representative training sample. We present a generic, extensible framework for learning the multi-dimensional representation of user intent from the query words. The approach models the latent relationships between facets using tree structured distribution which leads to an efficient and convergent algorithm, FastQ, for identifying the multi-faceted intent of users based on just the query words. We also incorporated WordNet to extend the system capabilities to queries which contain words that do not appear in the training data. Empirical results show that FastQ yields accurate identification of intent when compared to a gold standard.

Glioblastoma-Specific Protein Interaction Network Identifies PP1A and CSK21 as Connecting Molecules between Cell Cycle-Associated Genes

Glioblastoma (GBM; grade IV astrocytoma) is a very aggressive form of brain cancer with a poor survival and few qualified predictive markers. This study integrates experimentally validated genes that showed specific upregulation in GBM along with their protein-protein interaction information. A system level analysis was used to construct GBM-specific network. Computation of topological parameters of networks showed scale-free pattern and hierarchical organization. From the large network involving 1,447 proteins, we synthesized subnetworks and annotated them with highly enriched biological processes. A careful dissection of the functional modules, important nodes, and their connections identified two novel intermediary molecules CSK21 and protein phosphatase 1 alpha (PP1A) connecting the two subnetworks CDC2-PTEN-TOP2A-CAV1-P53 and CDC2-CAV1-RB-P53-PTEN, respectively. Real-time quantitative reverse transcription-PCR analysis revealed CSK21 to be moderately upregulated and PP1A to be overexpressed by 20-fold in GBM tumor samples. Immunohistochemical staining revealed nuclear expression of PP1A only in GBM samples. Thus, CSK21 and PP1A, whose functions are intimately associated with cell cycle regulation, might play key role in gliomagenesis. Cancer Res; 70(16); 6437-47. (C)2010 AACR.

Font and Size Independent OCR for Printed Kannada Documents using SVM Classifiers

The following topics were dealt with: document analysis and recognition; multimedia document processing; character recognition; document image processing; cheque processing; form processing; music processing; document segmentation; electronic documents; character classification; handwritten character recognition; information retrieval; postal automation; font recognition; Indian language OCR; handwriting recognition; performance evaluation; graphics recognition; oriental character recognition; and word recognition

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.

A post processing technique for improved harmonic/percussion separation in polyphonic speech

In this paper we propose a postprocessing technique for a spectrogram diffusion based harmonic/percussion decom- position algorithm. The proposed technique removes har- monic instrument leakages in the percussion enhanced out- puts of the baseline algorithm. The technique uses median filtering and an adaptive detection of percussive segments in subbands followed by piecewise signal reconstruction using envelope properties to ensure that percussion is enhanced while harmonic leakages are suppressed. A new binary mask is created for the percussion signal which upon applying on the original signal improves harmonic versus percussion separation. We compare our algorithm with two recent techniques and show that on a database of polyphonic Indian music, the postprocessing algorithm improves the harmonic versus percussion decomposition significantly.

A transient detection algorithm for audio using iterative analysis of STFT

We propose an iterative algorithm to detect transient segments in audio signals. Short time Fourier transform(STFT) is used to detect rapid local changes in the audio signal. The algorithm has two steps that iteratively - (a) calculate a function of the STFT and (b) build a transient signal. A dynamic thresholding scheme is used to locate the potential positions of transients in the signal. The iterative procedure ensures that genuine transients are built up while the localised spectral noise are suppressed by using an energy criterion. The extracted transient signal is later compared to a ground truth dataset. The algorithm performed well on two databases. On the EBU-SQAM database of monophonic sounds, the algorithm achieved an F-measure of 90% while on our database of polyphonic audio an F-measure of 91% was achieved. This technique is being used as a preprocessing step for a tempo analysis algorithm and a TSR (Transients + Sines + Residue) decomposition scheme.

Case for multiple views of function in design based on a common definition

Functions are important in designing. However, several issues hinder progress with the understanding and usage of functions: lack of a clear and overarching definition of function, lack of overall justifications for the inevitability of the multiple views of function, and scarcity of systematic attempts to relate these views with one another. To help resolve these, the objectives of this research are to propose a common definition of function that underlies the multiple views in literature and to identify and validate the views of function that are logically justified to be present in designing. Function is defined as a change intended by designers between two scenarios: before and after the introduction of the design. A framework is proposed that comprises the above definition of function and an empirically validated model of designing, extended generate, evaluate, modify, and select of state-change, and an action, part, phenomenon, input, organ, and effect model of causality (Known as GEMS of SAPPhIRE), comprising the views of activity, outcome, requirement-solution-information, and system-environment. The framework is used to identify the logically possible views of function in the context of designing and is validated by comparing these with the views of function in the literature. Describing the different views of function using the proposed framework should enable comparisons and determine relationships among the various views, leading to better understanding and usage of functions in designing.

Learning from positive and unlabelled examples using maximum margin clustering

Learning from Positive and Unlabelled examples (LPU) has emerged as an important problem in data mining and information retrieval applications. Existing techniques are not ideally suited for real world scenarios where the datasets are linearly inseparable, as they either build linear classifiers or the non-linear classifiers fail to achieve the desired performance. In this work, we propose to extend maximum margin clustering ideas and present an iterative procedure to design a non-linear classifier for LPU. In particular, we build a least squares support vector classifier, suitable for handling this problem due to symmetry of its loss function. Further, we present techniques for appropriately initializing the labels of unlabelled examples and for enforcing the ratio of positive to negative examples while obtaining these labels. Experiments on real-world datasets demonstrate that the non-linear classifier designed using the proposed approach gives significantly better generalization performance than the existing relevant approaches for LPU.