The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
We analyze online social data to model social interactions of users in recommender systems: i) Rating prediction, and ii) detecting spammers and abnormal user rating behaviors. We propose a social trust model using matrix factorization method to estimate users taste by incorporating user-item matrix. The effect of users friends tastes is modeled based on centrality metrics and similarity algorithms...
Modeling and predicting people's opinions plays an important role in today's life. For viral marketing and political strategy design, it is particularly important to be able to analyze competing opinions, such as pro-Democrat vs. pro-Republican. While observing the evolution of polar opinions in a social network over time, can we tell when the network "behaved"' abnormally? Furthermore,...
Various techniques have been proposed to detect smells in spreadsheets, which are susceptible to errors. These techniques typically detect spreadsheet smells through a mechanism based on a fixed set of patterns or metric thresholds. Unlike conventional programs, tabulation styles vary greatly across spreadsheets. Smell detection based on fixed patterns or metric thresholds, which are insensitive to...
A major challenge in Cloud computing is resource provisioning for computational tasks. Not surprisingly, previous work has established a number of solutions to provide Cloud resources in an efficient manner. However, in order to realize a holistic resource provisioning model, a prediction of the future resource consumption of upcoming computational tasks is necessary. Nevertheless, the topic of prediction...
In this paper, we explore the redundancy in convolutional neural network, which scales with the complexity of vision tasks. Considering that many front-end visual systems are interested in only a limited range of visual targets, the removing of task-specified network redundancy can promote a wide range of potential applications. We propose a task-specified knowledge distillation algorithm to derive...
Tagging of faces present in a photo or video at shot level has multiple applications related to indexing and retrieval. Face clustering, which aims to group similar faces corresponding to an individual, is a fundamental step of face tagging. We present a progressive method of applying easy-to-hard grouping technique that applies increasingly sophisticated feature descriptors and classifiers on reducing...
Pest camouflages in grains or natural environment cause significant difficulties in pest detection using imaging technologies. This paper proposes a convolutional Riemannian texture with differential entropic active contours to distinguish the background regions and expose pest regions. An image texture model is firstly introduced on the Riemannian manifold. A convolutional Riemannian texture structure...
In applications involving matching of image sets, the information from multiple images must be effectively exploited to represent each set. State-of-the-art methods use probabilistic distribution or subspace to model a set and use specific distance measure to compare two sets. These methods are slow to compute and not compact to use in a large scale scenario. Learning-based hashing is often used in...
Appearance based person re-identification in real-world video surveillance systems is a challenging problem for many reasons, including ineptness of existing low level features under significant viewpoint, illumination, or camera characteristic changes to robustly describe a person's appearance. One approach to handle appearance variability is to learn similarity metrics or ranking functions to implicitly...
This paper proposes a handling strategy for dynamic resource events in Cyber-Physical Production Systems (CPPS). Because of the increasing complexity and the dynamic run-time behavior, future automation systems become more and more fragile. New methods are needed to deal with such a dynamic behavior, exploiting existing degrees of freedom. Currently, existing methods are limited to single resource...
Community networks establish a wireless mesh network among citizens, providing a network that is independent, free, and (in some cases) available where regular Internet access is not. Following initial disappointments with their performance and availability, they are currently experiencing a second spring. Many of these networks are growing fast, but with little planning and limited oversight. Problems...
This paper describes a reliable methodology for radar cross section (RCS) measurement of complex small and large targets in the W band. The backscattering behavior of a small car model was measured in an anechoic chamber along with various automotive related targets in a wide gymnasium. Experimental performance in the anechoic chamber is compared to the simulation results. Our simulation model is...
A Convolutional Neural Networks (CNNs) approach is proposed to automate the method of Diabetic Retinopathy(DR) screening using color fundus retinal photography as input. Our network uses CNN along with denoising to identify features like micro-aneurysms and haemorrhages on the retina. Our models were developed leveraging Theano, an open source numerical computation library for Python. We trained this...
Cloud computing provides support for hosting client's application. Cloud is a distributed platform that provides hardware, software and network resources to both execute consumer's application and also to store and mange user's data. Cloud is also used to execute scientific workflow applications that are in general complex in nature when compared to other applications. Since cloud is a distributed...
Building program analysis tools is hard. A recurring development task is the implementation of the meta-model around which a tool is usually constructed. The XCORE prototype supports this task by generating the implementation of the meta-model. For this purpose, developers will add directly into the source code of the tool under construction some meta-information describing the desired meta-model...
Exploring the design space of the memory hierarchy requires the use of effective methodologies, tools, and models to evaluate different parameter values. Reuse distance is of one of the locality models used in the design exploration and permits analytical cache miss estimation, program characterization, and synthetic trace generation. Unfortunately, the reuse distance is limited to a single locality...
This paper shows that many applications exhibit execution-phase-specific sensitivity towards approximation of the internal subcomputations. Therefore, approximation in certain phases can be more beneficial than others. Further, this paper presents Opprox, a novel system for application's execution-phase-aware approximation. For a user provided error budget and target input parameters, Opprox identifies...
Just-In-Time (JIT) defect prediction models aim to predict the commits that will introduce defects in the future. Traditionally, JIT defect prediction models are trained using metrics that are primarily derived from aspects of the code change itself (e.g., the size of the change, the author’s prior experience). In addition to the code that is submitted during a commit, authors write commit messages,...
In this artifact, we partially address the problem of development of smart Cyber-Physical Systems (sCPS) by providing a concrete model problem and testbed for experimenting with, comparing, and developing new adaptation techniques and algorithms pertinent to sCPS. In particular, our model problem features autonomous robots cooperating opportunistically in a highly dynamic environment with multiple...
Collaborative Filtering technique is a recognized technique used in recommender systems for providing useful recommendations to users. The domain dependent nature of Collaborative Filtering allows more diverse set of recommendations at the same time making the user interested in recommendation process. Cold start problem is the most inevitable problem in collaborative filtering which makes it difficult...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.