Tweets and Gab Postings for source classification

Data can be open using pickle with "rb" model.<br>LanguageExample object could be found here:https://gist.github.com/billweasley/b8e05969b5b2267473d13f263253812e<br><br><br>equally_split_examples.pickle:<br><br>- Hashtag Equally Devide Dataset [Ensure Hashtag (topic) distribution]- It is a dictionary containing two keys: "train" and "dev"<br>- The value in the dictionary is a list of <i>LanguageExample </i>object- 1168 Tweets, 1168 Gab Postings- Label: "0" indicates the sample from Gab, "1" indicates the sample from Twitter<br>random_split_examples.pickle:<br><br>- Random Devide Dataset [Ensure only the gab and tweet has a same number of samples, but do not ensure it has same distribution of hashtags for the different social network platforms]- It is a dictionary containing two keys: "train" and "dev"<br>- The value in the dictionary is a list of <i>LanguageExample </i>object- 2181 Tweets, 2181 Gab Postings- Label: "0" indicates the sample from Gab, "1" indicates the sample from Twitter<br>gabs_tags_information_maps.pickle- Original Gabs collected and pre-processed (URL removels, HTTP tag removel, HTTP decoding)- A map where the key is hashtags and the value is a list of gab language associated to the correspoding hashtag.<br><br>tweets_tags_information_maps.pickle- Original Tweets collected and pre-processed (URL removels, HTTP tag removel, HTTP decoding)- A map where the key is hashtags and the value is a list of tweets language associated to the correspoding hashtag.