Financial market predictions utilize historical data to anticipate future stock prices and market trends. Traditionally, these predictions have focused on the statistical analysis of quantitative factors, such as stock prices, trading volumes, inflation rates, and changes in industrial production. Recent advancements in large language models motivate the integrated financial analysis of both sentiment data, particularly market news, and numerical factors. Nonetheless, this methodology frequently encounters constraints due to the paucity of extensive datasets that amalgamate both quantitative and qualitative sentiment analyses. To address this challenge, we introduce a large-scale financial dataset, namely, Financial News and Stock Price Integration Dataset (FNSPID). It comprises 29.7 million stock prices and 15.7 million time-aligned financial news records for 4,775 S&P500 companies, covering the period from 1999 to 2023, sourced from 4 stock market news websites. We demonstrate that FNSPID excels existing stock market datasets in scale and diversity while uniquely incorporating sentiment information. Through financial analysis experiments on FNSPID, we propose: (1) the dataset's size and quality significantly boost market prediction accuracy; (2) adding sentiment scores modestly enhances performance on the transformer-based model; (3) a reproducible procedure that can update the dataset. Completed work, code, documentation, and examples are available at this http URL. FNSPID offers unprecedented opportunities for the financial research community to advance predictive modeling and analysis.

RelKD 2023 Workshop at KDD 2023

Bloodstain pattern analysis plays a crucial role in crime scene investigations by providing valuable information through the study of unique blood patterns. Conventional image analysis methods, like Thresholding and Contrast, impose stringent requirements on the image background and is labor-intensive in the context of droplet image segmentation. The Segment Anything Model (SAM), a recently proposed method for extensive image recognition, is yet to be adequately assessed for its accuracy and efficiency on bloodstain image segmentation. This paper explores the application of pre-trained SAM and fine-tuned SAM on bloodstain image segmentation with diverse image backgrounds. Experiment results indicate that both pre-trained and fine-tuned SAM perform the bloodstain image segmentation task with satisfactory accuracy and efficiency, while fine-tuned SAM achieves an overall 2.2% accuracy improvement than pre-trained SAM and 4.70% acceleration in terms of speed for image recognition. Analysis of factors that influence bloodstain recognition is carried out. This research demonstrates the potential application of SAM on bloodstain image segmentation, showcasing the effectiveness of Artificial Intelligence application in criminology research. We release all code and demos at https://github.com/Zdong104/Bloodstain_Analysis_Ai_Tool

The 38th Annual AAAI Conference on Artificial Intelligence AAAI 2024

The rapid evolution of artificial intelligence (AI), specifically large language models (LLMs), has opened opportunities for various educational applications. This paper explored the feasibility of utilizing ChatGPT for automating feedback for Java programming assignments in an introductory computer science (CS1) class. The study employs surveys for 102 students responses, following formative feedback guidelines, and performed thematic analysis of open-ended responses. Specifically, this study focused on three questions: 1) To what extent do students view LLM-generated feedback as formative? 2) How do students see the comparative affordances of feedback prompts that include their code, vs. those that exclude it? 3) What enhancements do students suggest for improving AI-generated feedback? This study summarizes key findings, emphasizing the complexity of incorporating LLMs in education. demonstrated that LLMs could generate Java programming assignment feedback that students perceived as formative. It also offered insights into the specific improvements that would make the ChatGPT-generated feedback useful for students.

[ICAIBD 2023] The 6th International Conference on Artificial Intelligence and Big Data

We propose to augment ChatGPT with math performance assessments, which facilitate the creation of customized learning experiences based on the needs of each student. This study explores how ChatGPT personalizes the learning experience, how it can be augmented with math and physical performance, and how educators can ensure that the LLM algorithm is unbiased.

[ASA 2023] The Acoustic Society of America

As we previously reported in rodent lungs, RMT parameters (Expected value E(x), and , the eigenvalue with the highest probability) extracted from singular value decomposition of Inter-element Matrix (IRM) show a significant correlation with histology scores. Lack of fibrotic models for bigger lungs such as pigs encourages us to investigate 3D printed PEGDA phantoms with controllable pore size and thickness for further investigation. We hypothesize that E(x), and can distinguish with different phantoms and predict the backscattered signal regime dominancy. 128-element linear array L7-4 (Verasonics, at 5.2 MHz central frequency) connected to Verasonics Vantage. Phantoms of 1-inches size with different pore size of 0.085, 2*0.085, and 3*0.085 mm. Attenuation constants for these samples were measured by using Substitution Method at 1MHz, 2.25 MHz, and 5 MHz to confirm the results acquired from RMT. Both attenuation constants and RMT parameters show the larger the pore size is the more multiple scattering dominates the distribution of eigenvalues of the IRM. These preliminary results seem promising to use PEGDA phantoms to mimic pulmonary fibrosis, however further investigation need to confirm these results.

As an undergraduate student, I worked with my PhD student partner Mr. Christopher. I have been working on data processing using MATLAB and OpenSim to investigate the relationship between musculoskeletal structure and function in upper limb and hand motions. My work involves creating datasets and using MATLAB with OpenSim simulators to analyze and simulate data models. Additionally, I have been developing algorithms to optimize datasets and personalize bio-electric signals for hand muscle controls. To ensure that the experiment data is analyzed effectively, I have been processing the data using MATLAB and preparing the necessary data for publication. Through this project, I have gained valuable experience in data processing, algorithm development, and simulation modeling, which will undoubtedly contribute to my professional development as a researcher in the field of biomechanics.

I have worked on a project assigned by a PhD student, focusing on processing MRI images of porcine hindlimbs. Utilizing 3D segmentation tools, I performed segmentation for the epiphysis part of both the femur and tibia in ITK Snap4 and 3D Slicer5, and exported the segmentation as an stl. file in 3D Slicer to construct a 3D model. Notch width and femoral condyle width were measured for both the intact and ACL-injured knees, and statistical shape modeling was performed between the two legs using ShapeWorks. Through this work, I was able to compare the shape difference, bone angle difference, and function difference between the intact and ACL-injured legs in the porcine experiment.