Doctor of Philosophy (PhD), Ohio University, 1990, Electrical Engineering & Computer Science (Engineering and Technology)

This dissertation studies the theoretical underpinnings of active sonar classifiers. We present a systematic approach for designing optimal Bayesian classifiers and analyzing their performance. We emphasize the ternary case where three hypotheses are considered: H 0(noise only), H 1(reverberation plus noise) and H 2(target plus noise). We start by deriving a sufficient statistic to decide between H 1and H 2, assuming H 0has already been eliminated. Then, closed-form solutions for classification and false alarm probabilities are obtained and several receiver operating characteristics curves illustrating meaningful physical scenarios are presented. Two classes of illuminating signals are considered: high resolution and linear FM signals. Many design parameters affecting classifier performance are studied. Perhaps the most important issue is classifier performance when incorrect a priori knowledge of the target's spatial properties is processed. Other parameters such as target resolution, signal-to-noise ratio, transmitter constant in linear FM signals, etc. are investigated as well. The final issue presented is acoustic target imaging. A minimum variance linear unbiased estimator of the scattering coefficients of the test volume encompassing the target is derived. Furthermore, we investigate error minimization of the MVLU estimator in terms of system characteristics such as array and/or signal design. We also discuss the relation between classification and imaging. In summary, ideas from decision theory, detection and estimation theory are combined in order to implement optimal Bayesian classifiers and acoustic imagers.

Committee: John Tague (Advisor) Subjects:

Master of Science in Electrical Engineering, University of Dayton, 2024, Electrical Engineering

This thesis presents a novel method to control an industrial robotic arm using multiple sensors. This system consists of a hybrid brain activity and vision sensors that convey a human being's intention and visual perception. We fuse and analyze the data from those sensors using a machine learning-based approach to automatically guide the manipulator to a designated location. We believe that this Brain–Machine–Interface (BMI) can greatly alleviate the burdensome traditional method used to program a robot (greatly aids the end-user). We experiment with different modular configurations for the brain activity information, i.e., parallelized models and what we refer to as a global model for fusing the information. We explore various machine learning and pattern recognition techniques as well as existing feature selection methods. Our experimental results show that the subject can control the robot to a destination of interest using a machine—robot–interface. We attain accuracy in the order of 99.6% when it comes to the desired motion and 99.8% for the case of deducing the desired characteristic (color) of the targeted object. These results outperform any similar existing approaches that we have researched. Moreover, in comparison to those similar operational systems, we present a unique modular configuration for brain activity interpretation and object detection mechanism that yields an overall system that is highly computationally efficient. Although, in this work, we implemented and demoed our approach using a simple pick and place demo, our work presents the basic structure underlying a system that can be efficiently used to benefit people with restricted ability to function physically (tetraplegic patients), and allowing them to perform complex and robotics related duties in an industrial setting.

Committee: Temesguen Messay-Kebede (Advisor); Barath Narayanan (Committee Member); Russell Hardie (Committee Member) Subjects: Computer Engineering; Computer Science; Electrical Engineering; Engineering; Remote Sensing; Robotics

Master of Arts, Case Western Reserve University, 2024, Cognitive Linguistics

This corpus-based study investigates how adult L2 learners' L1 and proficiency level affect their usage of classifiers in Mandarin Chinese. The study focuses on learners from 6 different L1 groups (English, French, German, Japanese, Korean and Thai). Each token of the classifier is annotated as one of the following: (1) Correct, (2) Missing, (3) Overgeneralization of the general classifier 个 ge, (4) Misuse of a specific classifier, and (6) Other. The results show that (1) learners' L1 and proficiency levels are related to their probability of correctly using the classifier, (2) learners with different L1s tend to make different types of errors. Learners show better performance when referring to objects that are typical examples of the specific classifier, which is aligned with the prototype category theory. Based on the above results, a Cognitive Linguistics instructional material for three sets of Mandarin classifiers is designed and presented.

Committee: Yasuhiro Shirai (Committee Chair); Todd Oakley (Committee Member); Vera Tobin (Committee Member) Subjects: Linguistics

Master of Science, Miami University, 2023, Computer Science and Software Engineering

Software models are used to analyze and understand the properties of the system, providing stakeholders with an overview of how the system should work before actually implementing it. Such models are usually created informally, such as drawing sketches on a whiteboard or paper, especially during the early design phase, because these methods foster communication and collaboration among stakeholders. However, these informal sketches must be formalized to be useful in later applications, such as analysis, code generation, and documentation. This formalization process is often tedious, error-prone, and time-consuming. In an effort to avoid recreating formal models from scratch, this thesis presents SkeMo, a sketch-based software modeling tool. SkeMo is built on a CNN-based image classifier using 3000 input sketches of class diagram components and integrated into the functionality of an existing web-based model editor, the Instructional Modeling Language (IML), with a newly implemented touch interface. SkeMo was evaluated using a ten-fold cross-validation to assess the image classifier and through a user study involving 20 participants to collect metrics and feedback. The results demonstrate the promising potential of sketch-based modeling as an intuitive and efficient modeling practice, allowing users to quickly and easily create models to design complex software systems.

Committee: Eric Rapos (Advisor); Christopher Vendome (Committee Member); Xianglong Feng (Committee Member); Douglas Troy (Committee Member) Subjects: Computer Science; Engineering

Master of Science, University of Toledo, 2023, Cyber Security

With the rapid increase of social network spam, it's essential to empower users with the tools to detect the harmful spam effectively. However, existing tools cannot meet the requirements. In this paper, we propose and develop a live detection tool that can detect ham and spam text and images from social networks, this tool will be trained on user collected data (Image and Text) using different classifiers, where text and images are pre-processed and then passed onto the classifier that the user can choose, the user is then able to save the model and load it whenever they want to use a social network, where this tool will show the user a notification alerting them whether the post they are looking at is spam or ham before they even get the chance to read the text or look at the image, thus protecting them from clicking on malicious links that might harm their computer and steal their data. Evaluation results have demonstrated the effectiveness of our tool.

Committee: Weiqing Sun (Committee Chair); Hong Wang (Committee Member); Ahmad Javaid (Committee Member) Subjects: Computer Science

Master of Science, University of Toledo, 2018, Electrical Engineering

Real-time voice recognition and environmental sound detection play an important role in the fields of security, home control systems, robotics, and speech forensics. The advantages and its potential need in these industries have been a great motivation behind this work. The task of voice recognition and environmental sound detection is challenging due to high variability in sound signals. Furthermore, the presence of environmental noise makes the task of recognition even more difficult. Various methods and architectures have been introduced for both voice and sound recognition till date. However, due to some limitations in these architectures, we came up with two different architectures for both voice recognition and background sound detection. Through these architectures, we try to overcome the limitations seen in the previous architectures proposed by various researchers. In this work for environmental sound detection, we present a real-time method in which features are extracted using standard signal processing techniques and classification is done using the standard ML based classifier. The extracted features are time domain features like ZCR and STE and frequency domain features like SC, SR, and SF. The Pitch was determined using Average Magnitude Difference Function (AMDF). For the classification, we used some robust and accurate ML techniques like SVM, RF, and DNN. Similarly, for voice recognition, we present a novel pipelined real-time end-to-end voice recognition architecture that enhances the performance of voice recognition by exploiting the advantages of GF and CNN. This architecture has been developed to provide a voice-user interface and aid in voice-based authentication and integration with an existing NLP system. Gaining secure access to existing NLP systems also served as one of the primary goals. Initially, in this work, we identify challenges related to real-time voice recognition and highlight the up-to-date research in the fiel (open full item for complete abstract)

Committee: Vijay Devabhaktuni (Committee Chair); Ahmad Javaid (Committee Co-Chair); Richard Molyet (Committee Member) Subjects: Computer Engineering; Electrical Engineering

Master of Sciences (Engineering), Case Western Reserve University, 2017, EECS - Computer and Information Sciences

This work investigates a strategy for evaluating the navigability of terrain from 3-D imaging. Labeled training data was automatically generated by running a simulation of a mobile robot nai¨vely exploring a virtual world. During this exploration, sections of terrain were perceived through simulated depth imaging and saved with labels of safe or unsafe, depending on the outcome of the robot's experience driving through the perceived regions. This labeled data was used to train a deep convolutional neural network. Once trained, the network was able to evaluate the safety of perceived regions. The trained network was shown to be effective in achieving safe, autonomous driving through novel, challenging, unmapped terrain.

Committee: Wyatt Newman (Advisor); Cenk Cavusoglu (Committee Member); Michael Lewicki (Committee Member) Subjects: Computer Science; Robotics; Robots

PhD, University of Cincinnati, 2007, Engineering : Computer Science

This thesis proposes a fuzzy set - based classifier for imbalanced data sets, that is when one class, the majority class, or the data set provided for it, is much larger than the other class, the minority class. Current machine learning classification algorithms are biased to the majority class, and therefore perform poorly in recognition of the minority class. The experiments in this thesis show that the proposed classifier eliminates to a large extent this bias by considering a fuzzy set from frequency class representation that takes into account class size. In addition, it also analyzes the effect on the classifier of other characteristics of data such as overlap, complexity, and size, in combination with the imbalance factor. Capabilities and limitations of the proposed fuzzy classifiers are extensively investigated along a range of data sets that combine imbalance with the above factors. The relation of the proposed fuzzy classifier with another, often used frequency - based classifier, namely the Naive Bayes classifier, is considered. A theoretical result indicates that Naive Bayes classifier is a particular case of the fuzzy classifier presented here. More precisely, it is shown that the Bayes classification criterion, the Bayes score, can be obtained as a particular case of constructing the fuzzy set, and hence the fuzzy classifier. Finally, for cases where data re-balancing is necessary, e.g. extremely imbalanced data, an up-sampling algorithm that incorporates information about the whole data set, such as imbalance and distances between and within classes, is proposed.

Committee: Dr. Anca Ralescu (Advisor) Subjects: Computer Science