About the Internship

The Alliance is currently working on improving the accuracy of large language models to extract relevant information from scientific text. However, the agricultural research domain has a lower amount of annotated textual data, making it a low-resource domain. To ensure the AI models are as accurate as possible and to promote the reproducibility of our research, the Alliance needs to annotate a corpus of textual data for training and testing the models. We are seeking BSc or MSc level Economics students to assist in annotating textual data for a Named Entity Recognition (NER) project focused on economics. The data sources include scientific publications, reports, and news media.

Key Responsibilities

• Undergo training on Prodigy, a Python-based annotation tool
• Review and annotate assigned textual data in the agricultural research domain, identifying relevant named entities
• Apply consistent annotation guidelines across all data sources
• Participate in regular quality assurance checks to uphold high standards for data quality and annotation consistency.
• Contribute to research and analysis to identify trends and patterns in the annotated data.
• Report any issues or ambiguities encountered during the annotation process
• Meet weekly annotation targets as set by the project coordinator
• Contribute to the refinement of annotation guidelines based on practical experience

Requirements

• Current enrollment in or recent graduation from a BSc or MSc program in Economics, Agricultural Economics, Agricultural Sciences, Environmental Science, or related fields
• Strong analytical and research skills, with a keen eye for detail.
• Understanding of economic and agricultural concepts and terminology.
• Fluency in English and excellent communication and interpersonal skills.
• Basic computer literacy and ability to learn new software tools
• Exposure to academic research methodologies, knowledge of python or other statistical software will be an added advantage