Over 10,000 physical typewritten documents from 1932 to 1941 had to be digitised, structured, and connected in order to create a single, centralised source of knowledge, for enabling the analysis of historical processes.
Fabio Montagna is Lead Machine Learning Engineer at GraphAware and presented Temporal Graph Analysis at NODES2022. In this session, we’ll share our experience with horizon scanning over a graph of medical research papers. By leveraging the author keywords from scientific publications, it’s possible to build a cooccurrence graph with a temporal component provided by the paper publication date. We’ll show how we can analyze trends and evolution patterns using an unsupervised algorithm that assigns roles to author keyword.
Federica Ventruto and Alessia Melania Lonoce are Junior Data Scientists at GraphAware who spoke at NODES2022. Natural language processing is an indispensable toolkit to build knowledge graphs from unstructured data. However, it comes with a price. Keywords and entities in unstructured texts are ambiguous - the same concept can be expressed by many different linguistic variations. The resulting knowledge graph would thus be polluted with many nodes representing the same entity without any order. In this session, we show how the semantic similarity based on transformer embeddings and agglomerative clustering can help in the domain of academic disciplines and research fields and how Neo4j improves the browsing experience of this knowledge graph.
Vlasta Kůs is Lead Data Scientist at GraphAware and presented at NODES2022. Public archives contain incredible amount of knowledge. In this session, we’ll cover a real use case of building a knowledge graph for the archive of a major foundation to help empower researchers (or business analysts) to access previously unavailable levels of insights. This archive, going up to a century back, contains detailed information about funded projects and conversations preceding them, budgets, research endeavors, and outcomes, as well as priceless knowledge about influence networks of foundation representatives, researchers, and students. A particular challenge was that the same events were described in multiple sources. The only way to leverage all of this knowledge was through the use of advanced analytics and machine learning. We will explore the technologies (including OCR, NLP, and graph data science) and complex pipelines employed to create this major knowledge graph.
‘The amount of savings in time and effort [the search optimization] can deliver for our home offices, for our customers, is incredible.’
--Mayank Gupta, SVP for data, LPL Financial
GraphAware and Neo4j experts demonstrate how you can leverage knowledge graphs to help with compliance challenges.
‘GraphAware Hume and Neo4j have significantly reduced the amount of manual effort required to keep documentation consistent and mitigate compliance risk.’
Vlasta Kůs takes us through converting a corpus of research papers through Natural Language Processing, entity (relation) extraction and graph algorithms to highly informative connected insights organized in a knowledge graph.
Christophe Willemsen, CTO, GraphAware, explains how to apply NLP to extract entities and key phrases to build and search knowledge graphs
Mayank Gupta, SVP of Data and Wren Chan, VP of Foundational Architecture and Innovation from LPL Financial present how they use GraphAware Hume and Neo4j to power financial chat bots.
‘The ability to customise Hume Actions via Cypher queries provided ESA with flexibility to cover a range of use cases and customers.’