Preprocessing, Inverted Index, TF-IDF, Vbyte Compression and Decompression
Mark: 6/6
The Gold Standard, Lin Similarity, NPMI, LSA, Pearson Correlation
Mark: 7/7
Language Modelling in Hangman
Mark: 7/7
Automatic Fact Verification
Mark: 23/30 இдஇ
Comment: -- the paper is well-written and mostly understandable Although there are several grammatical errors that could be easily fixed. the paper provides related work and a short summary of the dataset. the design techniques are mostly well-motivated. It's a bit unclear how certain hyper-parameters were tuned or thresholds set (e.g. why only 8 sentences were selected, have the authors tried other number/combinations?). No references to Tables/Figures (hard to infer what do they illustrate). The paper slightly exceeds page limits. No error analysis of the final system