Under copyright Constraint(s) on Use: This work is protected by the U.S. Copyright Law (Title 17, U.S.C.). Use of this work beyond that allowed by "fair use" or any license applied to this work requires written permission of the copyright holder(s). Responsibility for obtaining permissions and any use and distribution of this work rests exclusively with the user and not the UC San Diego Library. Inquiries can be made to the UC San Diego Library program having custody of the work. Use: This work is available from the UC San Diego Library. This digital copy of the work is intended to support research, teaching, and private study.
Rights Holder and Contact
Allen, Jessica K.; Henry, Ramona T.; Kale, Amol T.; Michael, Garrett J.; Stickle, Matthew P.F.
Description
The process of operating a food-related business is complex and requires in-depth knowledge of many factors, including local policies/regulations, supply chains, sources of funding, and more. These complicated factors have made the business of food a very difficult field to work in, for both new and existing professionals. To help with this effort, a comprehensive food business knowledge graph and chart based user interface, Nourish, was created in an attempt to reduce the barrier of finding the right information to successfully operate in the food industry. The knowledge graph was developed by integrating multiple data sources to address each component of the food industry. These data sources included geographic, document, ontological and relational data. For geographic data, the knowledge graph utilized Arcgis online via Esri to pinpoint optimal business locations (based on an opportunity scoring calculation) which could be suggested to the end user. For the document data, document stores were created to provide users with funding information from various government institutions, such as the Small Business Administration (SBA) and the United States Department of Agriculture (USDA). For the ontological data, various foundational ontologies including FOODON and FIBO were integrated into a graph database. For relational data, the USDA Food Data Central database was also incorporated to better understand nutritional alternatives and increase accessibility to healthy options. To enable widespread access to the knowledge graph, Nourish was connected to Open AI large language model (llm) GPT-3.5, which provided a user-friendly way to query the knowledge graph. To transform user queries with GPT-3.5 into responses, the ReAct conversational agent chain was implemented using the Langchain framework as an interface. The agent was composed of several tools to address user inputs. For example, the location tool was used to suggest optimal locations for a business. Another tool queried the document indexes created on the document stores to address loan eligibility and general loan inquiries. The Nourish chatbot was displayed through a Dash app to facilitate conversation with the user. Overall, the Nourish chatbot effectively queried the integrated knowledge graph to provide users with personalized recommendations on vital information such as where to open their business, what loans they could apply for, and where they could find additional support. Research Data Curation Program, UC San Diego, La Jolla, 92093-0175 (https://lib.ucsd.edu/rdcp) This project relies on external software packages, modules/libraries, or programs, use of which may carry specific license requirements. Users should comply with any licenses specified within the contents of this project. Allen, Jessica K.; Henry, Ramona T.; Kale, Amol T.; Michael, Garrett J.; Stickle Matthew P.F. (2023). Nourish: A Knowledge-Driven Recommender System for Food Entrepreneurs. In Data Science & Engineering Master of Advanced Study (DSE MAS) Capstone Projects. UC San Diego Library Digital Collections. https://doi.org/10.6075/J01N81B7 Since the project revolved around building and utilizing a knowledge base, there is not necessarily a clean input and output relationship for data. Normally the output from one step directly fed into the input of another step. For example, a lot of data ended up being part of the knowledge base ultimately used by the chatbot application.
Type
dataset
Identifier
ark:/20775/bb7587797q
Language
English
Subject
Prompt engineering Ontology Langchain Large language model (LLM) Document indexing Nutritional analysis Capstone projects ArcGIS Cypher Recommender system Data Science & Engineering Master of Advanced Study (DSE MAS) Knowledge graph DSE MAS - 2023 Cohort
If you're wondering about permissions and what you can do with this item, a good starting point is the "rights information" on this page. See our terms of use for more tips.
Share your story
Has Calisphere helped you advance your research, complete a project, or find something meaningful? We'd love to hear about it; please send us a message.
