001_Rule-Based Models/1-3_Matching Specific Tokens with Specific Attributes.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"provenance":[],"authorship_tag":"ABX9TyMHkevBE7pzvhqToskIIGGc"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"}},"cells":[{"cell_type":"markdown","source":["**Import Required Libraries**"],"metadata":{"id":"B4h5szZfgLqo"}},{"cell_type":"code","execution_count":null,"metadata":{"id":"jJARgaMrhyDh"},"outputs":[],"source":["import spacy\n","from spacy.matcher import Matcher"]},{"cell_type":"markdown","source":["\n","\n","*   import spacy: Imports the spacy library, which is used for natural language processing.\n","*   from spacy.matcher import Matcher: Imports the Matcher class from spacy.matcher, which is used to find patterns in text.\n","\n"],"metadata":{"id":"DBPn6ONMgO7q"}},{"cell_type":"markdown","source":["**Load Spacy Model**"],"metadata":{"id":"eSeVasixgVvl"}},{"cell_type":"code","source":["nlp=spacy.load('en_core_web_sm')"],"metadata":{"id":"b-ADoL-6h9FH"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["nlp = spacy.load('en_core_web_sm'): Loads the small English language model from spacy. This model contains the necessary tools and data for processing English text, including tokenization, part-of-speech tagging, named entity recognition, and more."],"metadata":{"id":"jYQ-QzZigYtL"}},{"cell_type":"markdown","source":["**Create Matcher Object**"],"metadata":{"id":"CL8nI7u5gcB5"}},{"cell_type":"code","source":["matcher=Matcher(nlp.vocab)"],"metadata":{"id":"7-7EukCriAwe"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["matcher = Matcher(nlp.vocab): Creates a Matcher object, which will be used to find patterns in text. The matcher is initialized with the vocabulary of the nlp object."],"metadata":{"id":"P6vKYlEagfH7"}},{"cell_type":"markdown","source":["**Define Pattern**"],"metadata":{"id":"rdxLSdWigh5_"}},{"cell_type":"code","source":["pattern=[{'LOWER': 'machine'},{'LOWER': 'learning'}]"],"metadata":{"id":"_y_TU1CdiDs9"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["pattern = [{'LOWER': 'machine'}, {'LOWER': 'learning'}]: Defines a pattern to match in the text. The pattern is a list of dictionaries, where each dictionary specifies attributes of a token to match. In this case:\n","\n","\n","*   The first token should be the lowercase word 'machine' ('LOWER': 'machine').\n","*   The second token should be the lowercase word 'learning' ('LOWER': 'learning').\n","\n"],"metadata":{"id":"xo8Yd0pTgl7S"}},{"cell_type":"markdown","source":["**Add Pattern to Matcher**"],"metadata":{"id":"8p15jwN7gt2d"}},{"cell_type":"code","source":["matcher.add('ML_PATTERN', [pattern])"],"metadata":{"id":"3htoBXcPiOvT"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["matcher.add('ML_PATTERN', [pattern]): Adds the defined pattern to the matcher with the name 'ML_PATTERN'."],"metadata":{"id":"mh8uhuwWgxk2"}},{"cell_type":"markdown","source":["**Define Text to be Processed**"],"metadata":{"id":"AGfIPiUIgzuW"}},{"cell_type":"code","source":["text=\"I am interested in machine learning and deep learning.\""],"metadata":{"id":"33to3xwziSwT"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["text = \"I am interested in machine learning and deep learning.\": Defines the text in which to search for the pattern."],"metadata":{"id":"aF2HdlEQg4FU"}},{"cell_type":"markdown","source":["**Process Text with Spacy**"],"metadata":{"id":"7vwLemdtg6rF"}},{"cell_type":"code","source":["doc=nlp(text)"],"metadata":{"id":"w3kMBUFNiYce"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["doc = nlp(text): Processes the text using the spacy model, creating a doc object that contains the processed text, including tokens, part-of-speech tags, entities, and more."],"metadata":{"id":"cu6ATFE-hAaM"}},{"cell_type":"markdown","source":["**Find Matches**"],"metadata":{"id":"sRwY6H4DhDOm"}},{"cell_type":"code","source":["matches = matcher(doc)"],"metadata":{"id":"59GwFjGliasc"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["matches = matcher(doc): Uses the matcher to find all instances of the defined pattern in the processed text (doc). The matcher returns a list of match objects."],"metadata":{"id":"3hvm2shAhGFJ"}},{"cell_type":"markdown","source":["**Process and Print Matches**"],"metadata":{"id":"vFjJxiEGhISW"}},{"cell_type":"code","source":["for match_id,start,end in matches:\n","  matched_span=doc[start:end]\n","  print(matched_span.text)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"2kqVrXXTiigO","executionInfo":{"status":"ok","timestamp":1720512613736,"user_tz":-360,"elapsed":6,"user":{"displayName":"Mahiyat Tanzim","userId":"03004994999431640800"}},"outputId":"3aeadcec-0921-4c62-a59d-19b4bafe03b4"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["machine learning\n"]}]},{"cell_type":"markdown","source":["\n","\n","*   for match_id, start, end in matches: Iterates over the matches returned by the matcher. Each match is a tuple containing a match ID, the start position, and the end position of the matched span.\n","*   matched_span = doc[start:end]: Creates a span object from the doc using the start and end positions. This span object represents the matched text.\n","*   print(matched_span.text): Prints the text of the matched span.\n","\n","\n"],"metadata":{"id":"fcnw5kUzhMdt"}},{"cell_type":"markdown","source":["**Summary**<br>\n","This notebook processes a text using spacy to create a doc object. It then uses the Matcher class to find patterns in the text. Specifically, it looks for instances where the word 'machine' is immediately followed by the word 'learning', both in lowercase. The notebook iterates over the matches and prints the text of each matched span. In this example, the pattern is designed to match the phrase \"machine learning\" in the text."],"metadata":{"id":"_nEQ-cQMhV3v"}}]}