research

My current research is focused on developing methods which exploit unstructured user generated web information in the process of making inferences about the magnitude of real-life events. The core of my work is based on the fields of Data Mining and Statistical (Machine) Learning. Publications are listed on this page.

Mood of Nation uses more than half a million geolocated tweets on a daily basis to detect mood and affect trends in the UK population. We focus on four categories, namely Joy, Sadness, Anger and Fear and make comparable inferences for several regions in the United Kingdom (such as South, North England etc.).

Note that this work is still under way (i.e. there is no published material yet) and the website is in a BETA version.

Mood in the UK

Mood of Nation infers Mood and Affect scores for several regions in the UK based on Twitter content
 

This work is concentrated on a generalisation of the methods used to exploit Twitter content for nowcasting flu rates in the population. In particular, an additional case study is presented, where rainfall rates in several UK cities are inferred by geolocated tweets. Inferring precipitation figures forms a much harder problem given that such a quantity does not have a stable behaviour (especially in the UK).

Nowcasting Events from the Social Web with Statistical Learning (V. Lampos and N. Cristianini, in ACM TIST)
We present a general methodology for inferring the occurrence and magnitude of an event or phenomenon by exploring the rich amount of unstructured textual information on the social part of the web. Having geo-tagged user posts on the microblogging service of Twitter as our input data, we investigate two case studies...
 
A press release for our work is available here. This work has been featured in several mainstream or science media, including ScienceDaily, PhysOrg, BBC Radio and ITV.
 

Word Cloud with automatically extracted n-grams used to track rainfall rates from Twitter content. Font size is proportional to an n-gram's importance weight and flipped words take negative weights. A preprint of the paper is available here.

Flu Detector is a tool that uses the content of Twitter for nowcasting the level of flu-like illness in several UK regions. The applied methodology is presented in the following publications:

Tracking the flu pandemic by monitoring the Social Web (V. Lampos and N. Cristianini, in CIP 2010)
We report on a monitoring method to measure the prevalence of disease in a population by analysing the contents of social networking tools, such as Twitter. We compare our flu-score with data from the Health Protection Agency, obtaining on average a statistically significant linear correlation which is greater than 95%...

Flu Detector - Tracking Epidemics on Twitter (V. Lampos, T. De Bie and N. Cristianini, in ECML/PKDD 2010)
Flu Detector is an automated tool with a web interface for tracking the prevalence of Influenza-like Illness (ILI) in several regions of the United Kingdom using the contents of Twitter's microblogging service...
 

Flu Detector - Tracking Epidemics on Twitter

This work has been featured in or pointed by mainstream media about advances in technology and science such as MIT Technology Review, New Scientist, Communications of the ACM and New York Times.

    
 

Weather talk exploited the content of blogs and news articles as well as road traffic data available on the web to infer weather states in several locations in the UK. The applied methodology has been presented in my MSc Thesis:

Weather talk - extracting weather information by text mining (V. Lampos, MSc Thesis supervised by N. Cristianini, 2008)

Weather talk - extracting weather information by text mining

A web page that visualises some results of this work.