added sense2vec, standalone and fuzzy matching support

added new features:

• sense2vec support #26
• standalone support
• fuzzy matching support #24
• binarized word2vec model support

minor bug fixes

• original data was over-written by expanded data => prior knowledge was only inferred once
• "key_word" and "word_key" got mixed up for logging when verbose=True

resolved bug where original data was overwritten

Overwriting the original data disables re-inferring the original knowledge to the rest of the model.

added option for Standalone usage

0.7.4

added support for standalone usage

added wildcard operator * to compund words patterns

0.7.3

Merge pull request #28 from Pandora-Intelligence/davidberenstein1957-…

resolve python 3.8 versioning error

0.7.2

added typing from list to List

bug fix case sensitive entity scoring

#25 entity scoring did not work in combination with case sensitivity

bug fixes and efficiency updates by Tom Aarsen

• Resolve several breaking bugs.
  

• Perform a large optimization regarding string normalization and KeyedVector membership.
  

• Perform simple code optimization each localized to just a few lines.
  

small feature updates, bug fixes and documentation update

#17 resolved an issue that caused duplicate logging for the same missing keys. Also introduced verbose option.
#18 introduced json_path to export matching patterns to a custom path.
#19 defaults to key if example words are not present in the embedding model.

added more extensive logging and better default parameters

0.6.2

corrected for missing class initialization

resolved #13

Added correct character escaping Regex lowercase match
Added correct n-gram join.