CriticalMAAS MinMod

at USC-ISI

Critical minerals underpin modern tech and the global economy. MinMod uses AI to process diverse datasets, automating grade and tonnage model creation, expediting mineral site data updates, and aiding mine deposit location.

Project Website · Explorer

Linked Maps

at USC-ISI

Framework for extracting and semantically linking content from vast historical map archives. The initiative unlocks rich insights from maps, promoting interdisciplinary research in socio-geographic processes.

Project Website

Shbita, B. & Knoblock, C.A. (2024, Feb). Automatically Constructing a Geospatial Feature Taxonomy from OpenStreetMap Data. In Proceedings of the 18th International Conference on Semantic Computing (ICSC 2024), (pp. 208–211). IEEE

Link · Paper · Slides · Code

Shbita, B., Knoblock, C.A., Duan, W., Chiang, Y.Y., Uhl, J.H. & Leyk, S. (2023, April). Building Spatio-Temporal Knowledge Graphs from Vectorized Topographic Historical Maps. Semantic Web, 14(3), (pp. 527–549). IOS Press

Link · Paper · Slides · Code · Explorer

Shbita, B., Knoblock, C. A., Duan, W., Chiang, Y. Y., Uhl, J. H., & Leyk, S. (2020, May). Building Linked Spatio-Temporal Data from Vectorized Historical Maps. In European Semantic Web Conference (pp. 409-426). Springer, Cham

Link · Paper · Slides

Understanding Customer Requirements - an Enterprise Knowledge Graph Approach

with Dr. Anna Lisa Gentile, Dr. Pengyuan Li, Mr. Chad DeLuca and Dr. Guang-Jie Ren (IBM Research)

Leveraging linked open data and Enterprise Knowledge Graphs to guide the extraction and formalization of customer requirements from text.

Project Website

Shbita, B., Gentile, A. L., Li, P., DeLuca, C., & Ren, G. (2023). Understanding Customer Requirements-An Enterprise Knowledge Graph Approach. In Extended Semantic Web Conference, pages 625–643, Cham, Springer Nature Switzerland

Link · Paper · Slides

Automated Generation of Control Concepts Annotation Rules Using Inductive Logic Programming

with Dr. Abha Moitra (GE Research)

Automatically generating classification and annotation rules for control concepts in cyber-physical systems (CPS) using semantic technologies and Inductive Logic Programming.

Shbita, B., & Moitra, A. (2022). Automated Generation of Control Concepts Annotation Rules Using Inductive Logic Programming. In Hanus, M.; and Igarashi, A., editor(s), Functional and Logic Programming, pages 171–185, Cham, Springer International Publishing

Link · Paper · Slides · Talk (video)

Table Understanding

at USC-ISI

Extracting and Interpreting Time Series for Causal Discovery.

Project Website

Shbita, B., Rajendran, A., Pujara, J., & Knoblock, C. A. (2019). Parsing, Representing and Transforming Units of Measure. In Modeling the World’s Systems

Paper · Slides · Code · Code (API)

MINT (Model INTegration)

at USC-ISI

Scientific Model Integration through Knowledge-Rich Data and Process Composition.

Project Website

Shbita, B., Vu, B., Feldman, D., Pham, M., Rajendran, A., Knoblock, C. A., Pujara, J., & Chiang, Y. Y. (2019). Creating a FAIR Data Catalog to Support Scientific Modeling. In Workshop on Advanced Knowledge Technologies for Science in a FAIR World (AKTS)

Paper · Slides

Viola @ Socialbot Grand Challenge 4

semi-finalists at university challenge to advance human-computer interaction (Amazon Science)

An open-domain dialogue system for spoken conversation that uses a topic-agnostic dialogue manager based on a simple generate-and-rank approach.

Team Webpage · Video (find me @ 17:36)

H. Cho, B. Shbita, K. Shenoy, S. Liu, N. Patel, H. Pindikanti, J. Lee, and J. May. (2021). Viola: A Topic Agnostic Generate-and-Rank Dialogue System. In Proceedings of the 4th Alexa Prize

Link · Paper

Augmenting Conversational Dialogue Datasets with Commonsense and Adaptive Local Knowledge

with Hyundong Cho (USC)

Designed a dialogue dataset augmentation framework to expand multi-turn dialogue systems to incorporate external knowledge sources and ground pre-trained language models with commonsense reasoning.
Project conducted as part of the 'CSCI 662: Advanced Natural Language Processing' class.

Paper

Robust Defense Against Lp-Norm-Based Attacks by Learning Robust Representations

with Iordanis Fostiropoulos, and Myrl Marmarelis (USC)

Investigated the vulnerablity of image classifiers to adversarial perturbations and proposed a new regularization objective based on the Euclidean distance between classes in the hypothetical latent space found in the features of the penultimate layer.
Project conducted as part of the 'CSCI 699: Adversarial Machine Learning' class.

Paper

Music Latent Representation and Disentanglement for Genre Classification

with Ehsan Qasemi, Binh Vu, and Minh Pham (USC)

Investigated the use of unsupervised representation learning techniques to compress musical samples into a low-dimensional representation and use it for the task of music genre classification.
Project conducted as part of the 'CSCI 699: Representation Learning: Theory and Practice' class.

Paper

Marvel and DC-Comics comic-books Knowledge Graph

Supervisors: Prof. Pedro Szekely, Dr. Jay Pujara (USC)

Planned, designed, implemented and deployed a knowledge graph which captures Marvel and DC-Comics brands comic-book characters, comic-book issues and movies.
Project conducted as part of the 'INF 558/CSCI 563: Building Knowledge Graphs' class.

Paper · Slides · Demo (video) · Code

IPoIB Router in the Switch-X Product Family: from Firmware Design to Implementation of the Verification Environment and the Debug Tools

under direction of Mr. Ami Marelli (Mellanox Technologies / Nvidia)

Executed an implementation and verification process of the IPolB Router firmware mechanism in the Switch-X IC, a Switch System Silicon designed by Mellanox Technologies.
The development was done in the firmware layer using different methods and tools, according to the company’s required methodologies.

Physical Design of 12-bit Two's Complement Adder

Supervisors: Prof. Yosi Shacham, Mr. Avi Efrati (Tel Aviv University)

The project included hierarchical layout design, floorplanning, manual estimations and full layout simulations.
The Design was implemented using the 90nm Generic Process Design Kit (gpdk090) Technology and the gsclib090 library and was designed in Assura Virtuoso software.
Project conducted as part of the 'Introduction to VLSI design' course.

Logic Design of Full Simplified DLX using FPGA

Supervisors: Mrs. Liron David, Mr. Marko Markov (Tel Aviv University)

This project includes hardware design, implementation and debugging of a simple RISC processor (including data and control planes) on an FPGA development platform.
Used Xilinx software package and RESA software suite to code schematic and hardware description language (VHDL) designs and conduct logical simulations.
Project conducted as part of the 'Advanced Computer Structure Lab' course.

Kiosk Management: A Thesis in Systems Design and Programming with Web Services

Supervisor: Mrs. Inessa Bolotin (Israel Arts and Science Academy)

A user-friendly website for product purchasing.
The application features different levels of users and managers and enables them to view and add products to the catalog, prepare supplier-report and closely monitor expenses and revenues.
The system was used during senior high school year to raise money for students.

Improving Text-Independent Speaker Identification Performance Using Gaussian Mixture Speaker Models

under direction of Dr. David Palmer (HP Autonomy)

Studied Statistical Natural Language Processing, Automatic Speech Recognition and Machine Learning. Performed model training and testing with Python.
The research was conducted as part of the Research Science Institute (RSI), an intensive six-week program under sponsorship of the Center for Excellence in Education and hosted by the MIT.

Paper