Information Retrieval (IR) from structured data is facilitated by schemas such as ontologies, which enable efficient data access and organization. However, most knowledge is recorded in unstructured formats using natural language, which induces additional efforts to retrieve information. Retrieval Augmented Generation (RAG) has been developed to support general IR in unstructured text, yet it often produces inconsistencies due to model limitations. To address these inconsistencies, we propose transforming unstructured texts into structured information using text-to-Knowledge Graph (KG) translation. KGs support robust retrieval and reasoning tools, making them well-suited for complex knowledge representation. They are also understandable for humans. Our approach employs open-ended information extraction (for node extraction) and open-ended relation extraction (for relation extraction). To ensure the fidelity of the information present in the input text, we want to leverage a new combination of syntactic analysis for entity and relation extraction and Large Language Model (LLM) analysis to link the entities together. Initial implementations of this pipeline showed that our method effectively produces understandable and queryable KG.