The files consist of:
-
Project Proposal: Project proposal that contains a detailed step by step explanation of the process (old file).
-
Presentation Presentation of the project, proposed steps (old file).
-
00_webscrapping: Scrapping pages and parsing html to text.
-
01_imdb: Querying omdb API to retrive film meta data.
-
02_nlp: Use of NLTK library for tokenization, lemmation and text statistics.
-
03_watson: Watson API interface to perform personality insights on the scripts.
-
04_cleaning: Remove NA's, corrupted data, clean and order dataset.
-
05_preprocessing: Scale data to have mean=0 and std = 1, used for learning models.
-
06_visualization: Visualize relationships among the features of the datasets and clusterings.
-
07_classification: Perform classification of script Genre using predictor features.
-
08_regression: Peform regression of script imdbRating using predictor features.
-
09_recomendation: Recommend scripts based on their content and their rating.
-
10_summarization: Return script summary based on sentence words' frequencies.
-
11_word2vec: Create word embedings, word vectors using the whole script corpora, check word relationships.
-
data: Dataset in csv, Recommendation dictionary and word2vec files.
-
scripts: All the scrapped scripts.