Collection

2025 Northeast Section Conference

Authors

SUPARSHYA BABU SUKHAVASI; Thanu Sri Gandham; Susrutha Babu Sukhavasi; Meruva Veera Venkata Bhargav

to its three outputs. The input vector is denoted as I (A, B, C),potential to minimize power consumption and enhance while the output vector is represented as O (P, Q, R). Thecomputational efficiency. Numerous studies have examined relationship between the inputs and outputs follows specificthe role of reversible logic gates in the development of logical operations: P = A, Q = A ⊕ B (XOR operation), andenergy-efficient sequential circuits, leading to remarkable R = (A ⋅ B) ⊕ C (AND followed by XOR). Since the Peresprogress in digital circuit design. gate is reversible, it preserves information, making it highly In [1

Collection

2025 Northeast Section Conference

Authors

SUPARSHYA BABU SUKHAVASI; Susrutha Babu Sukhavasi; Mohammad Jaheerabi; Venkata Durga Sunanda Gangula

circuits, A B C P Q R P Q Rincluding the Toffoli and BJN gates, prevent such losses. 0 0 0 0 0 0 0 0 0 Quantum computing has benefited significantly from 0 0 1 0 0 1 0 0 1RLGs, where information is encoded in quantum states and 0 1 0 0 1 0 0 1 1 0 1 1 0 1 1 0 1 0 1 0 0 1 0 0 1 0 1 1 0 1 1 0 1 1 0 0 1 1 0 1 1 1 1 1 1 1 1 1 1 1 0 1 1 0

Collection

2025 Northeast Section Conference

Authors

Shruti Brahma, University of New Haven; Siddhant Alhat Rajendra, University of New Haven; Ardiana Sula, University of New Haven

-series statistical overview of the ARIMA approach and recurrent neural net-works (RNNs), specifically long-short-term memory (LSTM) TABLE Imodels are as follows. P ROJECTED G ROWTH R ATE S CENARIOS (OVERALL USA)A. Building Our Model: ARIMA and LSTM Growth Rate Type Value To create the ARIMA model, we first specify an (p, d, q) Average Growth Rate 0.0345 (3.45%)configuration and fit it to the enrollment data.The tuple (5,1,0) Maximum Growth Rate (Optimistic) 0.1146 (11.46%)represents

Collection

2025 Northeast Section Conference

Authors

Nusrat Zahan; Sidike Paheding

transformations, respectively. A. Quantitative AnalysisB. Model Analysis In the analysis of the UFO 120 and USR 248 datasets, data augmentation is found to be effective in enhancing PSNR and This study examines how data augmentation affects under- SSIM over models.water image super-resolution (SR) on three popular models,namely SCRNN, SRDRM, and DEEP SESR. SRCNN is TABLE Ihighly effective due to its simple and lightweight architec- Q UANTITATIVE EVALUATION OF DIFFERENT DATA AUGMENTATIONture, making it an ideal choice for IoT-based devices. Its

Collection

2025 Northeast Section Conference

Authors

Srilekha Bandla; Mukesh Reddy Jonnala; Peiqiao Wu; Sarosh Patel; Xingguo Xiong

by a stepper motor, ensuring controlled drugyet to create practical implementations that integrate CNN- delivery at a calculated rate:LSTM methodology with real-time therapeutic systems forpersonal treatment, which follows standard seizure detection Q = RP M × V (2)approaches described in [23] and [24]. An integrated system delivers VNS technology with a CNN- Where: - Q is the flow rate in mL/min, - RP M is the motorLSTM model and drug delivery systems as one functional unit. speed in rotations per minute, - V is the volume displaced perSeizures are detected through ECG motion data entry with revolution.real-time detection and adjustable intervention responses that

Collection

2025 Northeast Section Conference

Authors

Paul Cotae; Nian Zhang; Onyinye Obioha-Val

backscatter spectroscopy for standoff detection of trace explosives, “Optical Engineering, vol.59,no. 9,p.092009,2020.doi: 10.1117/1.OE.59.9.092009.[3] Chawla, Nitshe, Bowyer K, Hall L., Kegelmeyer W. “SMOTE: Synthetic Minority Over-Sampling Technique.” Journal of Artificial Research, vol.16, 2002, pp.321-357.[4] C.J. Breshike, C. A. Kendziora, R. Furstenberg, and R. A. McGrill, “Infrared backscatter imaging spectroscopy for standoff detection of trace explosive, “Journal of Applied Physics, vol.125, no.10, p.104901, 2019. doi:10.1063/1.5079622.[5] C.Liu, J. Li, M.E. Paoletti, J.M. Haut, A. Plaza and Q. Shi, “Accessibility-Free Active Learning for Hyperspectral Image Classification

Collection

2025 Northeast Section Conference

Authors

Shashi Kiran Chandrappa, Fairfield University; Sidike Paheding, Fairfield University

! 2n+1 X n X The Indian Pines dataset is a widely studied bench- f (x) = Φq ϕq,p (xp ) , (1)mark in HSI classification, captured using the Airborne q=1 p=1Visible/Infrared Imaging Spectrometer (AVIRIS) sensor. The where ϕq,p and Φq denote the univariate inner and outerdataset was collected in a region of agricultural fields near functions, respectively. In KANs, each of these inner functionsNorthwestern Indiana, USA, and is particularly

Collection

2025 Northeast Section Conference

Authors

Shashi Kiran Chandrappa; Sidike Paheding; Yu Cai Cai

delimiters This is awesome! - Response: Positiveand retrieving key attributes, including method, URL, headers, This is bad! - Response: Negativeand classification labels. Headers are dynamically mapped Wow that movie was good! - Response: Positiveinto a structured format, and categorical labels are assigned What a horrible show! -numerical values (Normal → 0, Anamoly → 1). A combined Model Output Output: Negative 3) Chain of thought: Detailed roadmap for your LLM’s rea-soning journey. It explicitly lays out each step the LLM shouldtake, from identifying the question to analyzing informationand drawing conclusions. Model Input Q: Roger has 5 tennis balls. He buys 2 more cans of tennis balls

Collection

2025 Northeast Section Conference

Authors

Safeena Khanam

. language understanding,” NAACL-HLT, 2019. • However, both methods failed to generate relevant [13] S. Reimers and I. Gurevych, “Sentence-BERT: Sentence embeddings using Siamese BERT-networks,” EMNLP, 2019. responses for Queries 4 and 5, suggesting that [14] Q. Zhu et al., “Efficient context retrieval in dialogue systems using neural certain queries may require alternative embedding similarity,” ACL, 2021. optimization strategies or more comprehensive [15] H. Jeon et al., “Memory-efficient transformer architectures for long- training data. context NLP,” ICLR

Collection

2025 Northeast Section Conference

Authors

Joseph P. Duszak; John F. Drazan; Cynthia A. Bautista

., vol. 48, no. 2, pp. 71–75, M. K. Muehlbauer, and A. Loomis, “Breaking Apr. 2023, doi: 10.1097/NNE.0000000000001318. Boundaries: How Immersive Virtual Reality Is[39] P. García-Pazo, S. Pol-Castañeda, C. Moreno-Mulet, A. physiological outcome measures,” Front. Virtual Real., Pomar-Forteza, and A. Carrero-Planells, “Virtual reality vol. 4, Aug. 2023, doi: 10.3389/frvir.2023.1211001. and critical care education in nursing: A cross-sectional [49] R. M. Dunnington, “Presence with Scenario-Based High study,” Nurse Educ. Today, vol. 131, p. 105971, Dec. Fidelity Human Patient Simulation,” Nurs. Sci. Q., vol. 2023, doi: 10.1016/j.nedt

Collection

2025 Northeast Section Conference

Authors

Saddam Alkhamaiesh; Peter Cavanugh

. Q. Patton, Qualitative Research and Evaluation Methods, 3rd ed. and Workforce Planning: Challenges and Best Practices.” Available: Thousand Oaks, CA: Sage, 2002. https://www.nrel.gov/transportation/ev-workforce.html [35] C. Cassell and G. Symon, Essential Guide to Qualitative Methods in[9] S. Alkhamaiesh, “Stakeholder Collaboration in EV Workforce Organizational Research, Thousand Oaks, CA: Sage, 2004. Training: A Thematic Analysis,” Proceedings of the American Society for Engineering Education (ASEE), 2024. [36] R. Yin, Case Study Research and Applications: Design and Methods

Collection

2025 Northeast Section Conference

Authors

Chushun Wang; A. Umit Coskun; Kai-tak Wan

expressed in terms of the heads due to hydrostatic pressure, flow rate, height difference, pump, turbine and dissipation [1]: Re: Renolds’ number defined by Re  Vd /  Q = (d2/4)V: volume flow rate in m3/s  P V2   P V2     z   hpump     z   hturbine  h f (1) g  g in g g hpump: head of pump in m

Collection

2025 Northeast Section Conference

Authors

Yegin Genc; Gonca Altuger-Genc; Akin Tatoglu

pipe has a diameter of 0.2 meters, and the flow rate [1] must be 100 liters per second (L/s). You need to calculate the required head self- the pump must provide, considering both the elevation difference and Interactive Learning Environments, vol. 31, no. 4, pp. 1974 1987, May frictional losses. 2023, Given: [2] -directed learning Flow rate, Q = 100 L/s = 0.1 m3/s Int J Educ

Collection

2025 Northeast Section Conference

Authors

Priscilla C. Fonseca; Riley Popp

concrete: helping computers to count air-voids like people count air-colors, results using RGB and HSV colorspace were nearly voids - methods for flatbed scanner calibration. Ph.D. thesis, Michiganidentical. For a variety of samples, some trial and error may Technological University (2008) [4] D. Zalocha, J. Kasperkiewicz, Cement and Concrete Research 35(10), 2041 (2005) [5] M. Radlinski, J. Olek, K.P. Q. Zhang, Concrete p. 64 (2010) [6] K.W. Peterson, R.A. Swartz, L.L. Sutter, T.J. Van Dam, in Transporta- tion Research Record, vol. 1775 (2001), vol. 1775, pp. 36–43 [7] L. Biehl, D. Landgrebe, Computers & Geosciences 28(10), 1153 (2002) [8] Multispec - a freeware multispectral image data analysis system

Collection

2025 Northeast Section Conference

Authors

Ahmed Hassebo; Kevin B. Montes

mechanisms, are planned. Future System Using Arduino UNO R3 and DHT11 Sensor," in 2020 17th International Computer Conference on Wavelet Active Mediaenhancements will focus on machine learning, weather Technology and Information Processing (ICCWAMTIP), Chengdu,forecasting, and solar-powered off-grid functionality. China, 2020.Future work will focus on enhancing system intelligence [9] Q. Qi and G. J. Brereton, "Mechanisms of removal of micron-sized particles by high-frequency

Collection

2025 Northeast Section Conference

Authors

Kalyan Khatry; Reihaneh Samsami

ethics geared towards engineers– covering topics like algorithmic bias, AI safety, and the socialimpacts of automation. This is crucial for future engineers REFERENCESwho will not only use AI but perhaps develop it or specify its [1] H. Li, T. Xu, C. Zhang, E. Chen, J. Liang, X. Fan, H. Li, J. Tang,use in projects. They must be equipped to ask the right and Q. Wen, “Bringing generative AI to adaptive learning inquestions: Is this AI output fair and safe? Is it respecting user education”, arXiv preprint arXiv:2402.14601, 2024.privacy? What are the

Collection

2025 Northeast Section Conference

Authors

Dinh Cuong Nguyen; Dan Tenney

Collection

2025 Northeast Section Conference

Authors

Jun Zhang; Peter Cavanaugh; Dan Tenney

Collection

2025 Northeast Section Conference

Authors

Nadia Albishi; Peter Cavanaugh