UMBC College of Engineering and Information Technology Dean's Office

Permanent URI for this collectionhttp://hdl.handle.net/11603/7919

Browse

Now showing 1 - 20 of 56

Qualitative Research Methods in Software Engineering: Past, Present, and Future
(IEEE, 2025) Seaman, Carolyn; Hoda, Rashina; Feldt, Robert
The paper entitled “Qualitative Methods in Empirical Studies of Software Engineering” by Carolyn Seaman was published in TSE in 1999. It has been chosen as one of the most influential papers from the third decade of TSE’s 50 years history. In this retrospective, the authors discuss the evolution of the use of qualitative methods in software engineering research, the impact it’s had on research and practice, and reflections on what is coming and deserves attention.
A LSTM with Dual-stage Attention Method to Predict Amine Emissions for Carbon Dioxide Capture and Storage
(IEEE, 2025-01-16) Rapelli, Sai Rajesh; Chen, Zhiyuan; Lu, Wei
To mitigate climate change impacts, carbon capture technologies have been implemented at significant CO2 emission points, such as industrial sites and electric power generation facilities. Solvent-based carbon capture solutions are pivotal in reducing atmospheric CO2 levels and enhancing air quality by capturing harmful pollutants. Amine-based solvents, favored for their efficiency in post-combustion CO2 capture, are susceptible to thermal and oxidative degradation, leading to complex emissions profiles that demand comprehensive management strategies. We develop a Machine Learning model designed to predict future amine emissions in real-time, thereby assisting in the formulation of mitigation strategies required for the operation of capture plants. We conducted an experiment using data from test campaigns run at the Technology Centre Mongstad (TCM). We employed a Long Short-Term Memory (LSTM) autoencoder model with dual-stage attention mechanisms to predict amine emissions using historical data. The results were quite promising: we achieved a mean absolute percentage error ranging from 5.8% to 6.8% percent for the real-time prediction of amine emissions. The results are better than existing approaches using simpler machine learning models as well as the standard LSTM autoencoder model.
Can Generative AI be Egalitarian?
(IEEE, 2024-10) Feldman, Philip; Foulds, James; Pan, Shimei
The recent explosion of “foundation” generative AI models has been built upon the extensive extraction of value from online sources, often without corresponding reciprocation. This pattern mirrors and intensifies the extractive practices of surveillance capitalism [46], while the potential for enormous profit has challenged technology organizations’ commitments to responsible AI practices, raising significant ethical and societal concerns. However, a promising alternative is emerging: the development of models that rely on content willingly and collaboratively provided by users. This article explores this “egalitarian” approach to generative AI, taking inspiration from the successful model of Wikipedia. We explore the potential implications of this approach for the design, development, and constraints of future foundation models. We argue that such an approach is not only ethically sound but may also lead to models that are more responsive to user needs, more diverse in their training data, and ultimately more aligned with societal values. Furthermore, we explore potential challenges and limitations of this approach, including issues of scalability, quality control, and potential biases inherent in volunteercontributed content.
Fair Inference for Discrete Latent Variable Models: An Intersectional Approach
(ACM, 2024-09-04) Islam, Rashidul; Pan, Shimei; Foulds, James
It is now widely acknowledged that machine learning models, trained on data without due care, often exhibit discriminatory behavior. Traditional fairness research has mainly focused on supervised learning tasks, particularly classification. While fairness in unsupervised learning has received some attention, the literature has primarily addressed fair representation learning of continuous embeddings. This paper, however, takes a different approach by investigating fairness in unsupervised learning using graphical models with discrete latent variables. We develop a fair stochastic variational inference method for discrete latent variables. Our approach uses a fairness penalty on the variational distribution that reflects the principles of intersectionality, a comprehensive perspective on fairness from the fields of law, social sciences, and humanities. Intersectional fairness brings the challenge of data sparsity in minibatches, which we address via a stochastic approximation approach. We first show the utility of our method in improving equity and fairness for clustering using naïve Bayes and Gaussian mixture models on benchmark datasets. To demonstrate the generality of our approach and its potential for real-world impact, we then develop a specialized graphical model for criminal justice risk assessments, and use our fairness approach to prevent the inferences from encoding unfair societal biases.
RAGged Edges: The Double-Edged Sword of Retrieval-Augmented Chatbots
(2024-06-12) Feldman, Philip; Foulds, James; Pan, Shimei
Large language models (LLMs) like ChatGPT demonstrate the remarkable progress of artificial intelligence. However, their tendency to hallucinate -- generate plausible but false information -- poses a significant challenge. This issue is critical, as seen in recent court cases where ChatGPT's use led to citations of non-existent legal rulings. This paper explores how Retrieval-Augmented Generation (RAG) can counter hallucinations by integrating external knowledge with prompts. We empirically evaluate RAG against standard LLMs using prompts designed to induce hallucinations. Our results show that RAG increases accuracy in some cases, but can still be misled when prompts directly contradict the model's pre-trained understanding. These findings highlight the complex nature of hallucinations and the need for more robust solutions to ensure LLM reliability in real-world applications. We offer practical recommendations for RAG deployment and discuss implications for the development of more trustworthy LLMs.
GenderAlign: An Alignment Dataset for Mitigating Gender Bias in Large Language Models
(2024-12-16) Zhang, Tao; Zeng, Ziqian; Xiao, Yuxiang; Zhuang, Huiping; Chen, Cen; Foulds, James; Pan, Shimei
Large Language Models (LLMs) are prone to generating content that exhibits gender biases, raising significant ethical concerns. Alignment, the process of fine-tuning LLMs to better align with desired behaviors, is recognized as an effective approach to mitigate gender biases. Although proprietary LLMs have made significant strides in mitigating gender bias, their alignment datasets are not publicly available. The commonly used and publicly available alignment dataset, HH-RLHF, still exhibits gender bias to some extent. There is a lack of publicly available alignment datasets specifically designed to address gender bias. Hence, we developed a new dataset named GenderAlign, aiming at mitigating a comprehensive set of gender biases in LLMs. This dataset comprises 8k single-turn dialogues, each paired with a "chosen" and a "rejected" response. Compared to the "rejected" responses, the "chosen" responses demonstrate lower levels of gender bias and higher quality. Furthermore, we categorized the gender biases in the "rejected" responses of GenderAlign into 4 principal categories. The experimental results show the effectiveness of GenderAlign in reducing gender bias in LLMs.
Foad Hamidi Launches New Projects To Expand Technology-rich Learning Opportunities For Youth In Baltimore
(UMBC News, 2024-12-17) Meyers, Catherine
Foad Hamidi, an assistant professor in the Department of Information Systems, has won funding from the National Science Foundation (NSF) to support two new projects offering technology-rich informal learning opportunities to diverse populations in Baltimore and beyond.
Mohamed Younis Honored For Contributions To Modern Communication Technologies
(UMBC News, 2024-12-13) Meyers, Catherine; Demond, Marlayna
Mohamed Younis, professor and chair of the Department of Computer Science and Electrical Engineering, has been honored by the Institute of Electrical and Electronics Engineers (IEEE) Communications Society for his significant and lasting contributions to the advancement of modern communication technologies. The award was announced December 9 at the society抯 Global Communications Conference in Cape...
What is the Point? Evaluating the Structure, Color, and Semantic Traits of Computer Vision Point Clouds of Vegetation
(MDPI, 2017-04-09) Dandois, Jonathan P.; Baker, Matthew; Olano, Marc; Parker, Geoffrey G.; Ellis, Erle C.
Remote sensing of the structural and spectral traits of vegetation is being transformed by structure from motion (SFM) algorithms that combine overlapping images to produce three-dimensional (3D) red-green-blue (RGB) point clouds. However, much remains unknown about how these point clouds are used to observe vegetation, limiting the understanding of the results and future applications. Here, we examine the content and quality of SFM point cloud 3D-RGB fusion observations. An SFM algorithm using the Scale Invariant Feature Transform (SIFT) feature detector was applied to create the 3D-RGB point clouds of a single tree and forest patches. The fusion quality was evaluated using targets placed within the tree and was compared to fusion measurements from terrestrial LIDAR (TLS). K-means clustering and manual classification were used to evaluate the semantic content of SIFT features. When targets were fully visible in the images, SFM assigned color in the correct place with a high accuracy (93%). The accuracy was lower when targets were shadowed or obscured (29%). Clustering and classification revealed that the SIFT features highlighted areas that were brighter or darker than their surroundings, showing little correspondence with canopy objects like leaves or branches, though the features showed some relationship to landscape context (e.g., canopy, pavement). Therefore, the results suggest that feature detectors play a critical role in determining how vegetation is sampled by SFM. Future research should consider developing feature detectors that are optimized for vegetation mapping, including extracting elements like leaves and flowers. Features should be considered the fundamental unit of SFM mapping, like the pixel in optical imaging and the laser pulse of LIDAR. Under optimal conditions, SFM fusion accuracy exceeded that of TLS, and the two systems produced similar representations of the overall tree shape. SFM is the lower-cost solution for obtaining accurate 3D-RGB fusion measurements of the outer surfaces of vegetation, the critical zone of interaction between vegetation, light, and the atmosphere from leaf to canopy scales.
Mitigating Demographic Bias in AI-based Resume Filtering
(ACM, 2020-07-13) Deshpande, Ketki V.; Pan, Shimei; Foulds, James
With increasing diversity in the labor market as well as the work force, employers receive resumes from an increasingly diverse population. However, studies and field experiments have confirmed the presence of bias in the labor market based on gender, race, and ethnicity. Many employers use automated resume screening to filter the many possible matches. Depending on how the automated screening algorithm is trained it can potentially exhibit bias towards a particular population by favoring certain socio-linguistic characteristics. The resume writing style and socio-linguistics are a potential source of bias as they correlate with protected characteristics such as ethnicity. A biased dataset is often translated into biased AI algorithms and de-biasing algorithms are being contemplated. In this work, we study the effects of socio-linguistic bias on resume to job description matching algorithms. We develop a simple technique, called fair-tf-idf, to match resumes with job descriptions in a fair way by mitigating the socio-linguistic bias.
Bayesian Modeling of Intersectional Fairness: The Variance of Bias
(SIAM, 2020-01) Foulds, James; Islam, Rashidul; Keya, Kamrun Naher; Pan, Shimei
Intersectionality is a framework that analyzes how interlocking systems of power and oppression affect individuals along overlapping dimensions including race, gender, sexual orientation, class, and disability. Intersectionality theory therefore implies it is important that fairness in artificial intelligence systems be protected with regard to multi-dimensional protected attributes. However, the measurement of fairness becomes statistically challenging in the multi-dimensional setting due to data sparsity, which increases rapidly in the number of dimensions, and in the values per dimension. We present a Bayesian probabilistic modeling approach for the reliable, data-efficient estimation of fairness with multidimensional protected attributes, which we apply to two existing intersectional fairness metrics. Experimental results on census data and the COMPAS criminal justice recidivism dataset demonstrate the utility of our methodology, and show that Bayesian methods are valuable for the modeling and measurement of fairness in intersectional contexts.
Fair Representation Learning for Heterogeneous Information Networks
(AAAI, 2021-05-22) Zeng, Ziqian; Islam, Rashidul; Keya, Kamrun Naher; Foulds, James; Song, Yangqiu; Pan, Shimei
Recently, much attention has been paid to the societal impact of AI, especially concerns regarding its fairness. A growing body of research has identified unfair AI systems and proposed methods to debias them, yet many challenges remain. Representation learning methods for Heterogeneous Information Networks (HINs), fundamental building blocks used in complex network mining, have socially consequential applications such as automated career counseling, but there have been few attempts to ensure that it will not encode or amplify harmful biases, e.g. sexism in the job market. To address this gap, we propose a comprehensive set of de-biasing methods for fair HINs representation learning, including sampling-based, projection-based, and graph neural networks (GNNs)-based techniques. We systematically study the behavior of these algorithms, especially their capability in balancing the trade-off between fairness and prediction accuracy. We evaluate the performance of the proposed methods in an automated career counseling application where we mitigate gender bias in career recommendation. Based on the evaluation results on two datasets, we identify the most effective fair HINs representation learning techniques under different conditions.
Can We Obtain Fairness For Free?
(ACM, 2021-07-30) Islam, Rashidul; Pan, Shimei; Foulds, James
There is growing awareness that AI and machine learning systems can in some cases learn to behave in unfair and discriminatory ways with harmful consequences. However, despite an enormous amount of research, techniques for ensuring AI fairness have yet to see widespread deployment in real systems. One of the main barriers is the conventional wisdom that fairness brings a cost in predictive performance metrics such as accuracy which could affect an organization's bottom-line. In this paper we take a closer look at this concern. Clearly fairness/performance trade-offs exist, but are they inevitable? In contrast to the conventional wisdom, we find that it is frequently possible, indeed straightforward, to improve on a trained model's fairness without sacrificing predictive performance. We systematically study the behavior of fair learning algorithms on a range of benchmark datasets, showing that it is possible to improve fairness to some degree with no loss (or even an improvement) in predictive performance via a sensible hyper-parameter selection strategy. Our results reveal a pathway toward increasing the deployment of fair AI methods, with potentially substantial positive real-world impacts.
ALDAS: Audio-Linguistic Data Augmentation for Spoofed Audio Detection
(2024-10-21) Khanjani, Zahra; Mallinson, Christine; Foulds, James; Janeja, Vandana
Spoofed audio, i.e. audio that is manipulated or AI-generated deepfake audio, is difficult to detect when only using acoustic features. Some recent innovative work involving AI-spoofed audio detection models augmented with phonetic and phonological features of spoken English, manually annotated by experts, led to improved model performance. While this augmented model produced substantial improvements over traditional acoustic features based models, a scalability challenge motivates inquiry into auto labeling of features. In this paper we propose an AI framework, Audio-Linguistic Data Augmentation for Spoofed audio detection (ALDAS), for auto labeling linguistic features. ALDAS is trained on linguistic features selected and extracted by sociolinguistics experts; these auto labeled features are used to evaluate the quality of ALDAS predictions. Findings indicate that while the detection enhancement is not as substantial as when involving the pure ground truth linguistic features, there is improvement in performance while achieving auto labeling. Labels generated by ALDAS are also validated by the sociolinguistics experts.
Creating Geospatial Trajectories from Human Trafficking Text Corpora
(2024-05-09) Karabatis, Saydeh N.; Janeja, Vandana
Human trafficking is a crime that affects the lives of millions of people across the globe. Traffickers exploit the victims through forced labor, involuntary sex, or organ harvesting. Migrant smuggling could also be seen as a form of human trafficking when the migrant fails to pay the smuggler and is forced into coerced activities. Several news agencies and anti-trafficking organizations have reported trafficking survivor stories that include the names of locations visited along the trafficking route. Identifying such routes can provide knowledge that is essential to preventing such heinous crimes. In this paper we propose a Narrative to Trajectory (N2T) information extraction system that analyzes reported narratives, extracts relevant information through the use of Natural Language Processing (NLP) techniques, and applies geospatial augmentation in order to automatically plot trajectories of human trafficking routes. We evaluate N2T on human trafficking text corpora and demonstrate that our approach of utilizing data preprocessing and augmenting database techniques with NLP libraries outperforms existing geolocation detection methods.
ALDAS: Audio-Linguistic Data Augmentation for Spoofed Audio Detection
(2024-10-21) Khanjani, Zahra; Mallinson, Christine; Foulds, James; Janeja, Vandana
Spoofed audio, i.e. audio that is manipulated or AI-generated deepfake audio, is difficult to detect when only using acoustic features. Some recent innovative work involving AI-spoofed audio detection models augmented with phonetic and phonological features of spoken English, manually annotated by experts, led to improved model performance. While this augmented model produced substantial improvements over traditional acoustic features based models, a scalability challenge motivates inquiry into auto labeling of features. In this paper we propose an AI framework, Audio-Linguistic Data Augmentation for Spoofed audio detection (ALDAS), for auto labeling linguistic features. ALDAS is trained on linguistic features selected and extracted by sociolinguistics experts; these auto labeled features are used to evaluate the quality of ALDAS predictions. Findings indicate that while the detection enhancement is not as substantial as when involving the pure ground truth linguistic features, there is improvement in performance while achieving auto labeling. Labels generated by ALDAS are also validated by the sociolinguistics experts.
Striving For More Efficient And Equitable Healthcare: Ian Stockwell Wins Major NIH Grant
(UMBC News, 2024-11-12) Meyers, Catherine
Let Students Take the Wheel: Introducing Post-Quantum Cryptography with Active Learning
(2024-10-17) Jamshidi, Ainaz; Kaur, Khushdeep; Gangopadhyay, Aryya; Zhang, Lei
Quantum computing presents a double-edged sword: while it has the potential to revolutionize fields such as artificial intelligence, optimization, healthcare, and so on, it simultaneously poses a threat to current cryptographic systems, such as public-key encryption. To address this threat, post-quantum cryptography (PQC) has been identified as the solution to secure existing software systems, promoting a national initiative to prepare the next generation with the necessary knowledge and skills. However, PQC is an emerging interdisciplinary topic, presenting significant challenges for educators and learners. This research proposes a novel active learning approach and assesses the best practices for teaching PQC to undergraduate and graduate students in the discipline of information systems. Our contributions are two-fold. First, we compare two instructional methods: 1) traditional faculty-led lectures and 2) student-led seminars, both integrated with active learning techniques such as hands-on coding exercises and Kahoot games. The effectiveness of these methods is evaluated through student assessments and surveys. Second, we have published our lecture video, slides, and findings so that other researchers and educators can reuse the courseware and materials to develop their own PQC learning modules. We employ statistical analysis (e.g., t-test and chi-square test) to compare the learning outcomes and students' feedback between the two learning methods in each course. Our findings suggest that student-led seminars significantly enhance learning outcomes, particularly for graduate students, where a notable improvement in comprehension and engagement is observed. Moving forward, we aim to scale these modules to diverse educational contexts and explore additional active learning and experiential learning strategies for teaching complex concepts of quantum information science.
Hearing the Voice of Software Practitioners on Technical Debt Monitoring: Understanding Monitoring Practices and the Practices' Avoidance Reasons
(Brazilian Computing Society, 2024-08-30) Freire, Sávio; Rios, Nicolli; Pérez, Boris; Castellanos, Camilo; Correal, Darío; Ramač, Robert; Mandić, Vladimir; Taušan, Nebojša; López, Gustavo; Pacheco, Alexia; Mendonça, Manoel; Falessi, Davide; Izurieta, Clemente; Seaman, Carolyn; Spínola, Rodrigo
Context. Technical debt (TD) monitoring allows software professionals to track the evolution of debt incurred in their projects. The technical literature has listed several practices used in the software industry to monitor indebtedness. However, there is limited evidence on the use and on the reasons to avoid using these practices. Aims. This work aims to investigate, from the point of view of software practitioners, the practices used for monitoring TD items, and the practice avoidance reasons (PARs) curbing the monitoring of TD items. Method. We analyze quantitatively and qualitatively a set of 653 answers collected with a family of industrial surveys distributed in six countries. Results. Practitioners are prone to monitor TD items, revealing 46 practices for monitoring the debt and 35 PARs for explaining TD non-monitoring. Both practices and PARs are strongly associated with planning and management issues. The study also shows the relationship found among practices, PARs and types of debt and presents a conceptual map that relates practices and PARs with their categories. Conclusion. The results of this study add to a practitioners’ capability to monitor TD items by revealing the monitoring practices, PARs and their relationship with different TD types.
Flood-ResNet50: Optimized Deep Learning Model for Efficient Flood Detection on Edge Device
(IEEE, 2024-03-19) Khan, Md Azim; Ahmed, Nadeem; Padela, Joyce; Raza, Muhammad Shehrose; Gangopadhyay, Aryya; Wang, Jianwu; Foulds, James; Busart, Carl; Erbacher, Robert F.
Floods are highly destructive natural disasters that result in significant economic losses and endanger human and wildlife lives. Efficiently monitoring Flooded areas through the utilization of deep learning models can contribute to mitigating these risks. This study focuses on the deployment of deep learning models specifically designed for classifying flooded and non-flooded in UAV images. In consideration of computational costs, we propose modified version of ResNet50 called Flood-ResNet50. By incorporating additional layers and leveraging transfer learning techniques, Flood-ResNet50 achieves comparable performance to larger models like VGG16/19, AlexNet, DenseNet161, EfficientNetB7, Swin(small), and vision transformer. Experimental results demonstrate that the proposed modification of ResNet50, incorporating additional layers, achieves a classification accuracy of 96.43%, F1 score of 86.36%, Recall of 81.11%, Precision of 92.41 %, model size 98MB and FLOPs 4.3 billions for the FloodNet dataset. When deployed on edge devices such as the Jetson Nano, our model demonstrates faster inference speed (820 ms), higher throughput (39.02 fps), and lower average power consumption (6.9 W) compared to larger ResNet101 and ResNet152 models.

Browse

Recent Submissions