NLP Natural Language Processing, Top Exam Review. 100% Accurate, verified.

NLP Natural Language Processing, Top Exam Review. 100% Accurate, verified. Stemming - -finding and comparing roots of words Data types - -unstructured and structured structured Dataset - -fixed dimensions (fixed number of rows and columns), well organised, tabular data, key-value pairs unstructured Dataset - -no fixed dimensions, no structure, can take any forms, audios, videos, images and text can be unstructured data. Information presented in unstructured data is not available for any analysis. key-value pairs (KVP) - -means there is one key, which is an unique identifier and a value, which is either the data or a pointer to the location of that data text data - -example of unstructured data, that can be Social Media: tweets, posts, comments; Conversations: messages, emails. chats; Articles: news and blogs, transcripts. Contains words arranged in a meaningful manner; is written in from of language and is defined by grammar and other structure natural language processing (NLP) - -is part of computer science and artificial intelligence which deals with human languages to gain information and insights. Applied NLP - -The use of NLP for designing and developing applications or systems in which an interaction between machines and natural languages Textmining - -derives useful information from text for Information and Insights. sentiment analysis - -a technique that allows marketers to analyze data from social media sites to collect consumer comments about companies and their products Applications of NLP - -Sentimental Analysis, Speech Recognition, Chatbot function, Machine Translation (Google Translate), Spell Checking, Keyword Searching, Advertising Matching, Information Extraction Natural Language Understanding - -Getting computers to understand and use natural language. Takes a lot of time and data Natural Language Generation - -Technology that enables computers to produce ordinary language by "voice" or on the screen so that people can understand computers more easily. (also voice synthesis) Tokenization - -First Step in NLP : parsing text into pieces of words and sentences Lemmatization - -grouping words together based on their basic dictionary definition (gone, going, went = go) POS tagging - -Grammarly matching and indication of sentences. Problems in differentiation (Google can be a noun or verb and a name ) Named entity recognition - -Identifies and extracts classes of entities (e.g. people, brands, firms, products, locations) and related information Chunking - -Picking up peaces and group them to meaning NLTK - -Natural Language Toolkit from Python to work