Learning Binary Code Representations for Effective and Efficient Image Retrieval

The size of online image datasets is constantly increasing. Considering an image dataset with millions of images, image retrieval becomes a seemingly intractable problem for exhaustive similarity search algorithms. Hashing methods, which encodes high-dimensional descriptors into compact binary strings, have become very popular because of their high efficiency in search and storage capacity. In the first part, we propose a multimodal retrieval method based on latent feature models. The procedure consists of a nonparametric Bayesian framework for learning underlying semantically meaningful abstract features in a multimodal dataset, a probabilistic retrieval model that allows cross-modal queries and an extension model for relevance feedback. In the second part, we focus on supervised hashing with kernels. We describe a flexible hashing procedure that treats binary codes and pairwise semantic similarity as latent and observed variables, respectively, in a probabilistic model based on Gaussian processes for binary classification. We present a scalable inference algorithm with the sparse pseudo-input Gaussian process (SPGP) model and distributed computing. In the last part, we define an incremental hashing strategy for dynamic databases where new images are added to the databases frequently. The method is based on a two-stage classification framework using binary and multi-class SVMs. The proposed method also enforces balance in binary codes by an imbalance penalty to obtain higher quality binary codes. We learn hash functions by an efficient algorithm where the NP-hard problem of finding optimal binary codes is solved via cyclic coordinate descent and SVMs are trained in a parallelized incremental manner. For modifications like adding images from an unseen class, we propose an incremental procedure for effective and efficient updates to the previous hash functions. Experiments on three large-scale image datasets demonstrate that the incremental strategy is capable of efficiently updating hash functions to the same retrieval performance as hashing from scratch.

Toxicity and Contamination in Bear Creek Sediment: Spatial Analysis and Implications for Risk Assessment

The sediments of Bear Creek near Baltimore, Maryland demonstrate substantial toxicity to benthic organisms, and contain a complex mixture of organic and inorganic contaminants. The present study maps the spatial extent and depth profile of toxicity and contamination in Bear Creek, and explores correlations between heavy metals, organic contaminants, and toxic responses. Two novel analytical techniques – handheld XRF and an antibody-based PAH biosensor – were applied to samples from the site to quantify total metals and total PAHs in sediments. By comprehensively assessing toxicity in Bear Creek, the present study provides data to inform future risk assessments and management decisions relating for the site, while demonstrating the benefits of applying joint biological assays and chemical assessment methods to sediments with complex contaminant mixtures.

User Behavioral Modeling of Web-based Systems for Continuous User Authentication

Authentication plays an important role in how we interact with computers, mobile devices, the web, etc. The idea of authentication is to uniquely identify a user before granting access to system privileges. For example, in recent years more corporate information and applications have been accessible via the Internet and Intranet. Many employees are working from remote locations and need access to secure corporate files. During this time, it is possible for malicious or unauthorized users to gain access to the system. For this reason, it is logical to have some mechanism in place to detect whether the logged-in user is the same user in control of the user's session. Therefore, highly secure authentication methods must be used. We posit that each of us is unique in our use of computer systems. It is this uniqueness that is leveraged to "continuously authenticate users" while they use web software. To monitor user behavior, n-gram models are used to capture user interactions with web-based software. This statistical language model essentially captures sequences and sub-sequences of user actions, their orderings, and temporal relationships that make them unique by providing a model of how each user typically behaves. Users are then continuously monitored during software operations. Large deviations from "normal behavior" can possibly indicate malicious or unintended behavior. This approach is implemented in a system called Intruder Detector (ID) that models user actions as embodied in web logs generated in response to a user's actions. User identification through web logs is cost-effective and non-intrusive. We perform experiments on a large fielded system with web logs of approximately 4000 users. For these experiments, we use two classification techniques; binary and multi-class classification. We evaluate model-specific differences of user behavior based on coarse-grain (i.e., role) and fine-grain (i.e., individual) analysis. A specific set of metrics are used to provide valuable insight into how each model performs. Intruder Detector achieves accurate results when identifying legitimate users and user types. This tool is also able to detect outliers in role-based user behavior with optimal performance. In addition to web applications, this continuous monitoring technique can be used with other user-based systems such as mobile devices and the analysis of network traffic.

ASSESSING THE LEVEL OF TECHNOLOGY INTEGRATION OF TITLE I TEACHERS IN A LARGE URBAN SCHOOL DISTRICT

By law, Title I schools employ teachers who are both competent in their subject knowledge and State certified. In addition, Title I teachers receive ongoing professional development in technology integration and are equipped with the latest innovative resources to integrate technology in the classroom. The aim is higher academic achievement and the effective use of technology in the classroom. The investment to implement technology in this large urban school district to improve student achievement has continued to increase. In order to infuse current and emerging technology throughout the curriculum, this school district needs to know where teachers have, and have not, integrated technology. Yet the level of how technology is integrated in Title I schools is unknown. This study used the Digital-Age Survey Levels of Teaching Innovation (LoTi) to assess 508 Title I teachers’ technology integration levels using three major initiatives purchased by Title I— the iPads program, the Chromebook initiative, and the interactive whiteboards program. The study used a quantitative approach. Descriptive statistics, regression analysis, and statistical correlations were used to examine the relationship between the level of technology integration and the following dependent variables: personal computer use (PCU), current instructional practices (CIP), and levels of teaching innovation (LoTi). With this information, budgetary decisions and professional development can be tailored to the meet the technology implementation needs of this district. The result of this study determined a significant relationship between the level of teaching innovation, personal computer use, and current instructional practices with teachers who teach with iPad, Chromebook, and/or interactive whiteboard. There was an increase in LoTi, PCU, and CIP scores with increasing years of experience of Title I teachers. There was also a significant relationship between teachers with 20 years or more teaching experience and their LoTi score.