(1) Background reading over the past few weeks.
(2) From now until the early September work on gathering (and classifying) a set of documents that can be used to test the accuracy of different sentiment extraction methods.Ê During this time I also intend to implement a baseline sentiment extraction system based on proximity-weighting, to compare against my implementation/s.
(3) Through September explore and implement potential NLP-analysis approaches for sentiment mining to improve on baseline system.
(4) Leave about 3 weeks to analyse results and prepare the report and presentation.