Scroll for more
events Sep 28, 2016

Text Analytics Meet-up

Hosted by



Sep 28, 2016


06:30 PM


Join London Text Analytics meet-up for an event on text analytics applied to biomedical relation extraction and to social media.

Biomedical relation extraction towards knowledge graph completion 
Claudiu Mihăilă (Data Scientist, BenevolentAI) 

The majority of biological knowledge resides in text (e.g., journal publications and patents), which has been undergoing exponential growth in the past few years. Although some of this information has been manually curated into structured databases, the process of curation is expensive and slow, and is unable to scale to the rate at which new knowledge is being produced. This necessitates the use of machine reading methods for automatic knowledge extraction. 
In this talk, we will describe the problem of biomedical relation extraction and knowledge graph completion, survey the state of the art, and discuss recent work in this domain.


Text Analytics Applied in Social Media Listening for Customer Insights 
Michalis Michael (CEO, DigitalMR) 

There are over 500 social media monitoring tools but only a handful may be appropriate for market research purposes. There are 5 main reasons for this: 

• 80-90% Irrelevant posts - Noise

• Lower than 60% sentiment accuracy for only one or a few languages

• No detection of emotions

• Semantic analysis is search based not inductive

• No integration with surveys and other traditional market research 

The audience will learn: 

- how these issues are now resolved,  

- how the extracted insights are used by organisations 

- why social listening and analytics is only 2% of the 65B US$ market research industry 

- what comes next in this space