Blog Archives

Partridge at BCS AI 2013

During the week of the 9th December 2013, we attended the BCS AI conference at Cambridge university to present Partridge to other members of the BCS Specialist Group for Artificial Intelligence (SGAI). The conference was held in the the very

Posted in Uncategorized Tagged with: , , , , ,

New Annotation Backend

Over the last couple of months I’ve been working with Dr Liakata on rewriting the SAPIENTA project in Python. The work was fun and rewarding and I got to visit Warwick University and deliver a talk about Partridge to some

Posted in Uncategorized Tagged with: , , , ,

BCS Paper

Over the last few weeks, my supervisors have been helping me in writing a short 6 page scientific paper for submission to the BCS SGAI Conference. This is hugely exciting since its my first ever scientific publication and if our

Posted in Uncategorized Tagged with: , , , , , , ,

New paper processing system

After a brief discussion with my supervisors, it turns out that Partridge has been incorrectly annotating papers (very very slightly) and this may have caused problems with our data integrety. That meant that we’re having to rebuild the database from

Posted in Uncategorized Tagged with: , , , , , , ,

Paper Type Analysis with Random Decision Forests

After long weeks of trying, I’ve managed to get some positive results from a paper type classifier that uses Random Decision Forests to classify papers into categories “Research Paper”, “Review Paper” or “Case Study”. Acquiring Test Data Firstly, I needed

Posted in Uncategorized Tagged with: , , , , , , ,

PlosGet.py – Downloading test data from PlosONE

One of the big problems that I’ve been having recently is a severe lack of test data for testing new machine learning behaviours with. I started off with just papers from the ART Corpus and manually cherrypicked some papers from

Posted in Uncategorized Tagged with: , , , , , , ,

Clustering adventures (contd.)

So I have been back through and done some more work on the clustering of the data. It looks like my algorithm is accurately clustering research and review papers correctly. To begin with  I had a very low k-means sillhouette

Posted in Uncategorized Tagged with: , , , , , ,

Using CoreSCs to determine a scientific paper’s ‘type’

Since Partridge’s very basic web interface went online earlier this week, I have been focussing my attention on how to make the system do more intelligent filtering operations. My first area of concern is “How can we tell what sort

Posted in Uncategorized Tagged with: , , , , , ,

Partridge now open for public testing.

The last couple of months have gone by incredibly quickly. I was minding my own business some time in December and before I knew it it, it was February! I’ve spent the last two weeks catching up on Partridge work

Posted in Uncategorized Tagged with: , , ,