Abstract

Search queries follow a long tail distribution which results in harder management of ad space for sponsored search. During keyword auctions, advertisers also tend to target head query keywords, thereby creating an imbalance in demand for head and tail keywords. This leads to under-utilization of ad space of tail query keywords. In this paper, we have explored a mechanism that allows the advertisers to bid on concepts rather than keywords. The tail query keywords are utilized by allocating a mix of head and tail keywords related to the concept. In the literature, an effort has been made to improve sponsored search by extracting the knowledge of coverage patterns among the keywords of transactional query logs. In this paper, we propose an improved approach to allow advertisers to bid on high level concepts instead of keywords in sponsored search. The proposed approach utilizes the knowledge of levelwise coverage patterns to allocate incoming search queries to advertisers in an efficient manner by utilizing the long tail. Experimental results on AOL search query data set show improvement in ad space utilization and reach of advertisers.

Abstract

Query recommendation (QR) support search engine to provide alternative queries as a recommendation using similarity-based approaches. In the literature, orthogonal query recommendation (OQR) has been proposed to compute the diversity of QR when the user does not formulate proper queries. The OQR uses dissimilarity measure in QR to recommend completely different queries. In this paper, we propose an approach in QR by extending association rules, diverse patterns, and unbalanced concept hierarchy of search terms. We conceptualize association rules based QR, and order the rules based on confidence and diversity. Subsequently, the high ranked rules based on confidence and diversity are provided in QRs. The experimental results on real world AOL click-through dataset show that the diverse QRs improve the performance significantly

Abstract

This paper describes the International Institute of Information Technology of Hyderabad’s submission to the task Classification Of Spanish Election Tweets (COSET) as a part of IBEREVAL-2017[1]. The task is to classify Spanish election tweets into political, policy, personal, campaign and other issues. Our system uses Support Vector Machines with radial basis function kernel to classify tweets. We dwell upon the character and word level features along with the word embeddings and train the classification model with them and present the results. Our best run achieves a F1-macro score of 0.6054 on the test corpus for first phase and 0.8509 for the second phase.

Abstract

The architectural trend towards heterogeneity has pushed heterogeneous computing to the fore of parallel computing research. Heterogeneous algorithms, often carefully handcrafted, have been designed for several important problems from parallel computing such as sorting, graph algorithms, matrix computations, and the like. A majority of these algorithms follow a work partitioning approach where the input is divided into appropriate sized parts so that individual devices can process the “right-sized” parts of the input. However, arriving at a good work partitioning is usually non-trivial and may crucially depend also on the input instance apart from the heterogeneous algorithm used. An extensive empirical search is not helpful as such an empirical search can potentially offset any gains accrued out of heterogeneous algorithms. Other recently proposed approaches too are in general inadequate. In this paper, we propose a simple and effective technique for work partitioning in the context of heterogeneous algorithms. Our technique is based on randomized sampling and therefore can adapt to both the algorithm used and the input instance. Our technique is generic in its applicability as we will demonstrate in this paper. We validate our technique on three problems: multiplying two unstructured sparse matrices (spmm), multiplying two scale-free sparse matrices, and finding the connected components of a graph (CC). For these problems, we show that using our method, we can find the required threshold that is under 10% away from the best possible thresholds respectively.

Abstract

What makes a computational neuronal model ’large scale’ ? Is it the number of neurons modeled? Or the number of brain regions modeled in a network? Most of the higher cognitive processes span across co-ordinated activity in a network of different brain areas. However at the same time, the basic information transfer takes place at a single neuron level, together with multiple other neurons. We explore modeling a neural system involving some areas of cortex, the basal ganglia (BG) and thalamus for the process of decision making, using a largescale neural engineering framework, Nengo. Early results tend to replicate the known neural activity patterns as found in the previous action selection model by Guthrie et al. 2013, besides operating with a larger neuronal populations. The power of converting algorithms to efficiently weighed neural networks in Nengo (Stewart et al. 2009 and Bekolay et al. 2013) is exploited in this work. Crucial aspects in a computational model, like parameter tuning and detailed neural implementations, while moving from a simplistic to large-scale model, are studied.

Abstract

We assess change in metastability to characterize age-effects on the dynamic repertoire of the functional networks at rest. Resting state fMRI signals from each subject (N=48) have been used and metastability is evaluated as the standard deviation of mean phase synchrony of BOLD signals across whole-brain as well as across known resting state networks. The results suggest that significant whole-brain metastability changes occur between middle to old age. We also demonstrate that static time-averaged FC largely undermines age-effects on the interaction between functional networks. Discriminant Function Analysis reveals existence of two different patterns of change in metastability, which maximally discriminates between two different processes of maturation and ageing.

Abstract

In this research paper, noise suppression ability of associative memories (e.g. Hopfield neural network) is briefly summarized. Motivated by this fact, noise suppression abilities of trained convolutional network is discussed. Introducing the concept of null vectors, noise suppression ability of Single Layer Perceptron, Multi-Layer Perceptron and Extreme Learning Machine are discussed.

Abstract

In the aftermath of 2015 Gorkha Earthquake in Nepal the challenging task was to classify structural damage; for further consideration of the building for living purposes. Manual classification of such a large number of building stock is tedious job in terms of collecting and maintaining data, particularly from the point of view of using the same for further planning purposes. Hence a web based application will make the job simple and easy for the surveyor as they will fill the details from the application and get the damage state of the building and know whether or not the building is safe for living and maintain the record of their decision. The other important feature of this application is it also does the pre-earthquake assessment of the building; thus it evaluates the state of the building before an earthquake through which occupants will know whether it safe or not to live in a particular building and also the improvements required from overall safety point of view. In this paper, description of the web application is given with some case studies. Post-earthquake assessment is done for two types of buildings which is quite common i.e., Reinforced Concrete Building and Brick Masonry Building. Pre-earthquake assessment is done in three parts, first part is Detail Visual Assessment of building based on BMTPC form and second part is Simplified Quantitative Assessment and third one is Detail Qualitative Assessment

Abstract

Mobile image retrieval allows users to identify visual information about their environment by transmitting image queries to an online image database that has associated annotations (e.g. location, product information, etc.) with the images. However, this is reliant on a network connection to transmit the query and retrieve the information. This chapter examines mobile image retrieval for offline use when the data net- work connection is limited or not available. In this scenario, the entire visual search index must reside on the mobile device itself. More specifically, we are interested in ‘instance retrieval’, where the annotations associated with the images (e.g. building’s name, object information) are returned by the query and not the images themselves. Figure 1 shows an example use case where mobile camera photos are used to identify buildings and landmarks without the need for a network connection. While our targeted instance retrieval does not need to store the images, the entire visual index structure needs to fit on a mobile device, ideally within a small footprint (e.g. 16–32 MB). This small memory footprint serves two purposes. First, while mobile phones have up to 16–32 GB of storage, this is mainly in the form of slower flash memory that is an order of magnitude slower than RAM. Having the entire index within tens of MBs makes it possible for use in a resident application on the phone’s RAM. Second, this small size is inline with common practices for mobile applications; e.g. the iPhone average application size is currently 23 MB. Addition- ally, iPhone apps less than 50 MB can be downloaded using 3G/4G, anything larger

Abstract

Building bilateral semantic associations between images and texts is among the fundamental problems in computer vision. In this paper, we study two complementary cross-modal prediction tasks: (i) predicting text(s) given a query image (“Im2Text”), and (ii) predicting image(s) given a piece of text (“Text2Im”). We make no assumption on the specific form of text; i.e., it could be either a set of labels, phrases, or even captions. We pose both these tasks in a retrieval framework. For Im2Text, given a query image, our goal is to retrieve a ranked list of semantically relevant texts from an independent text-corpus (i.e., texts with no corresponding images). Similarly, for Text2Im, given a query text, we aim to retrieve a ranked list of semantically relevant images from a collection of unannotated images (i.e., images without any associated textual meta-data). We propose a novel Structural SVM based unified framework for these two tasks, and show how it can be efficiently trained and tested. Using a variety of loss functions, extensive experiments are conducted on three popular datasets (two medium-scale datasets containing few thousands of samples, and one web-scale dataset containing one million samples). Experiments demonstrate that our framework gives promising results compared to competing baseline cross-modal search techniques, thus confirming its efficacy. Keywords: Image search; Image description; Cross-media analysis