This is a report, where preliminary work regarding the topic of voting intention inference from Social Media — such as Twitter — is presented. Our case study is the UK 2010 General Election and we are focusing on predicting the percentages of voting intention polls (conducted by YouGov) for the three major political parties — Conservatives, Labours and Liberal Democrats — during a 5-month period before the election date (May 6, 2010). We form three methodologies for extracting positive or negative sentiment from tweets, which build on each other, and then propose two supervised models for turning sentiment into voting intention percentages. Interestingly, when the content of tweets is enriched by attaching synonymous words, a significant improvement on inference performance is achieved reaching a mean absolute error of 4.34% ± 2.13%; in that case, the predictions are also shown to be statistically significant. The presented methods should be considered as work-in-progress; limitations and suggestions for future work appear in the final section of this script.