Cheating Detection In Online Examinations, 2015 SJSU
Cheating Detection In Online Examinations, Gaurav Kasliwal
In this research, we develop and analyze a tool that monitor student browsing activity during online examination. Our goal is to detect cheating in real time. In our design, a server capture packets using KISMET and detects cheating based on either a whitelist or blacklist of URLs. We provide implementation details and give experimental results, and we analyze various attack strategies. Finally, we show that the system is practical and lightweight in comparison to other available tools.
A Scalable Search Engine Aggregator, 2015 SJSU
A Scalable Search Engine Aggregator, Pooja Mishra
The ability to display different media sources in an appropriate way is an integral part of search engines such as Google, Yahoo, and Bing, as well as social networking sites like Facebook, etc. This project explores and implements various media-updating features of the open source search engine Yioop . These include news aggregation, video conversion and email distribution. An older, preexisting news update feature of Yioop was modified and scaled so that it can work on many machines. We redesigned and modified the user interface associated with a distributed news updater feature in Yioop. This project also introduced a video ...
An Open Source Advertisement Server, 2015 SJSU
An Open Source Advertisement Server, Pushkar Umaranikar
This report describes a new online advertisement system and its implementation for the Yioop open source search engine. This system was implemented for my CS298 project. It supports both selling advertisements and displaying them within search results. The selling of advertisement is done using a novel auction system, which we describe in this paper. With this auction system, it is possible to create an advertisement, attach keywords to it, and add it to the advertisement inventory. An advertisement is displayed on a search results page if the search keyword matches the keywords attached to the advertisement. Display of advertisements is ...
Context-Based Autosuggest On Graph Data, 2015 SJSU
Context-Based Autosuggest On Graph Data, Hai Nguyen
Autosuggest is an important feature in any search applications. Currently, most applications only suggest a single term based on how frequent that term appears in the indexed documents or how often it is searched upon. These approaches might not provide the most relevant suggestions because users often enter a series of related query terms to answer a question they have in mind. In this project, we implemented the Smart Solr Suggester plugin using a context-based approach that takes into account the relationships among search keywords. In particular, we used the keywords that the user has chosen so far in the ...
Ndex Strategies For Efficient And Effective Entity Search, Huy T. Vu
The volume of structured data has rapidly grown in recent years, when data-entity emerged as an abstraction that captures almost every data pieces. As a result, searching for a desired piece of information on the web could be a challenge in term of time and relevancy because the number of matching entities could be very large for a given query. This project concerns with the efficiency and effectiveness of such entity queries. The work contains two major parts: implement inverted indexing strategies so that queries can be searched in minimal time, and rank results based on features that are independent ...
Driver Telematics Analysis, 2015 SJSU
Driver Telematics Analysis, Karthik Vakati
For automobile insurance firms, telemetric analysis represents a valuable and growing way to identify the risk associated with each driver. The pricing decisions of an insurer are best accounted for if they are made considering the driver’s behavior instead of just the vehicle characteristics and the best way to understand a driver’s behavior is to leverage the telemetric analysis. Decisions made on such factors can eventually lead to increased premium or reduced liability for unsafe or reckless drivers and can also help in transitioning the burden to the policies that lead to increased liability.
The dataset provided for ...
Maximizing The Speed Of Influence In Social Networks, Yubo Wang
Influence maximization in social networks is the problem of selecting a limited
size of influential users as seed nodes so that the influence from these seed nodes can propagate to the largest number of other nodes in the network. Previous studies in influence maximization focused on three areas, i.e., designing propagation models, improving algorithms of seed-node selection and exploiting the structure of social networks. However, most of these studies ignored the time constraint in influence propagation. In this paper, I studied how to maximize influence propagation in a given time, i.e., maximizing the speed of influence propagation in ...
Using Neural Networks For Image Classification, Tim Kang
This paper will focus on applying neural network machine learning methods to images for the purpose of automatic detection and classification. The main advantage of using neural network methods in this project is its adeptness at fitting nonlinear data and its ability to work as an unsupervised algorithm. The algorithms will be run on common, publically available datasets, namely the MNIST and CIFAR10, so that our results will be easily reproducible.
Static Analysis Of Malicious Java Applets, 2015 SJSU
Static Analysis Of Malicious Java Applets, Nikitha Ganesh
In this research, we consider the problem of detecting malicious Java applets, based on static analysis. In general, dynamic analysis is more informative, but static analysis is more efficient, and hence more practical. Consequently, static analysis is preferred, provided we can obtain results comparable to those obtained using dynamic analysis. We conducted experiments with the machine learning technique, Hidden Markov Model (HMM). We show that in some cases a static technique can detect malicious Java applets with greater accuracy than previously published research that relied on dynamic analysis.
Combining Dynamic And Static Analysis For Malware Detection, Anusha Damodaran
Well-designed malware can evade static detection techniques, such as signature scanning. Dynamic analysis strips away one layer of obfuscation and hence such an approach can potentially provide more accurate detection results. However, dynamic analysis is generally more costly than static analysis. In this research, we analyze the effectiveness of using dynamic analysis to enhance the training phase, while using only static techniques in the detection phase. Relative to a fully static approach, the additional overhead is minimal, since training is essentially one-time work.
Video Event Understanding With Pattern Theory, 2015 University of South Florida
Video Event Understanding With Pattern Theory, Fillipe Souza, Sudeep Sarkar, Anuj Srivastava, Jingyong Su
We propose a combinatorial approach built on Grenander’s pattern theory to generate semantic interpretations of video events of human activities. The basic units of representations, termed generators, are linked with each other using pairwise connections, termed bonds, that satisfy predefined relations. Different generators are specified for different levels, from (image) features at the bottom level to (human) actions at the highest, providing a rich representation of items in a scene. The resulting configurations of connected generators provide scene interpretations; the inference goal is to parse given video data and generate high-probability configurations. The probabilistic structures are imposed using energies ...
Two Correspondence Problems Easier Than One, 2015 Purdue University
Two Correspondence Problems Easier Than One, Aaron Michaux, Zygmunt Pizlo
Computer vision research rarely makes use of symmetry in stereo reconstruction despite its established importance in perceptual psychology. Such stereo reconstructions produce visually satisfying figures with precisely located points and lines, even when input images have low or moderate resolution. However, because few invariants exist, there are no known general approaches to solving symmetry correspondence on real images. The problem is significantly easier when combined with the binocular correspondence problem, because each correspondence problem provides strong non-overlapping constraints on the solution space. We demonstrate a system that leverages these constraints to produce accurate stereo models from pairs of binocular images ...
Formal Aspects Of Non-Rigid-Shape-From-Motion Perception, 2015 SUNY College of Optometry
Formal Aspects Of Non-Rigid-Shape-From-Motion Perception, Vicky Froyen, Qasim Zaidi
Our world is full of objects that deform over time, for example animals, trees and clouds. Yet, the human visual system seems to readily disentangle object motions from non-rigid deformations, in order to categorize objects, recognize the nature of actions such as running or jumping, and even to infer intentions. A large body of experimental work has been devoted to extracting rigid structure from motion, but there is little experimental work on the perception of non-rigid 3-D shapes from motion (e.g. Jain, 2011). Similarly, until recently, almost all formal work had concentrated on the rigid case. In the last ...
Object Recognition And Visual Search With A Physiologically Grounded Model Of Visual Attention, 2015 Chemnitz University of Technology
Object Recognition And Visual Search With A Physiologically Grounded Model Of Visual Attention, Frederik Beuth, Fred H. Hamker
Visual attention models can explain a rich set of physiological data (Reynolds & Heeger, 2009, Neuron), but can rarely link these findings to real-world tasks. Here, we would like to narrow this gap with a novel, physiologically grounded model of visual attention by demonstrating its objects recognition abilities in noisy scenes.
To base the model on physiological data, we used a recently developed microcircuit model of visual attention (Beuth & Hamker, in revision, Vision Res) which explains a large set of attention experiments, e.g. biased competition, modulation of contrast response functions, tuning curves, and surround suppression. Objects are represented by object-view specific neurons, learned via a trace learning approach (Antonelli et al., 2014, IEEE TAMD). A visual cortex model combines the microcircuit with neuroanatomical properties like top-down attentional processing, hierarchical-increasing receptive field sizes, and synaptic transmission delays. The visual cortex model is complemented by a model of the frontal eye field (Zirnsak et al., 2011, Eur J Neurosci).
We evaluated the model on a realistic object recognition task in which a given target has to be localized in a scene (guided visual search task), using 100 different target objects, 1000 scenes, and two backgrounds. The model achieves an accuracy of 92% at black, and of 71% at white-noise backgrounds. We found that two of the underlying, neuronal attention mechanisms are prominently relevant for guided visual search: amplification of neurons preferring the target; and suppression of neurons encoding distractors or background noise.
Modeling Visual Features To Recognize Biological Motion: A Developmental Approach, 2015 Robotics, Brain and Cognitive Sciences Department, Istituto Italiano di Tecnologia, Italy
Modeling Visual Features To Recognize Biological Motion: A Developmental Approach, Giulio Sandini, Nicoletta Noceti, Alessia Vignolo, Alessandra Sciutti, Francesco Rea, Alessandro Verri, Francesca Odone
In this work we deal with the problem of designing and developing computational vision models – comparable to the early stages of the human development – using coarse low-level information.
More specifically, we consider a binary classification setting to characterize biological movements with respect to non-biological dynamic events. To this purpose, our model builds on top of the optical flow estimation, and abstract the representation to simulate the limited amount of visual information available at birth. We take inspiration from known biological motion regularities explained by the Two-Thirds Power Law, and design a motion representation that includes different low-level features, which can ...
Operational Semantics For Featherweight Lua, 2015 SJSU
Operational Semantics For Featherweight Lua, Hanshu Lin
Lua is a small, embedded language to provide scripting in other languages. De- spite a clean, minimal syntax, it is still too complex for formal reasoning because of some syntactic sugar or specific syntax structures in Lua.
∙ First-class ...
Using Probabilistic Graphical Models To Solve Np-Complete Puzzle Problems, Fengjiao Wu
Probabilistic Graphical Models (PGMs) are commonly used in machine learning to solve problems stemming from medicine, meteorology, speech recognition, image processing, intelligent tutoring, gambling, games, and biology. PGMs are applicable for both directed graph and undirected graph. In this work, I focus on the undirected graphical model. The objective of this work is to study how PGMs can be applied to find solutions to two puzzle problems, sudoku and jigsaw puzzles. First, both puzzle problems are represented as undirected graphs, and then I map the relations of nodes to PGMs and Belief Propagation (BP). This work represents the puzzle grid ...
Using Hidden Markov Models To Detect Dna Motifs, Santrupti Nerli
During the process of gene expression in eukaryotes, mRNA splicing is one of the key processes carried out by a complex called spliceosome. Spliceosome guarantees proper removal of introns and joining of exons before the translation process. Precise splicing is essential for the production of functional proteins. Spliceosome detects specific sequence motifs within an mRNA sequence called splice sites. Two of the splice sites are the 5’ and 3’ sites that border all the introns. Normal splicing process if disrupted by mutation may lead to fatal diseases. In this work, we predict splice sites in a human genome using hidden ...
Comparative Analysis Of Particle Swarm Optimization Algorithms For Text Feature Selection, Shuang Wu
With the rapid growth of Internet, more and more natural language text documents are available in electronic format, making automated text categorization a must in most fields. Due to the high dimensionality of text categorization tasks, feature selection is needed before executing document classification. There are basically two kinds of feature selection approaches: the filter approach and the wrapper approach. For the wrapper approach, a search algorithm for feature subsets and an evaluation algorithm for assessing the fitness of the selected feature subset are required. In this work, I focus on the comparison between two wrapper approaches. These two approaches ...