Text Processing

Tags

text-processing

The creation, storage and access of (massive) text in digital form by a computer.

Broca's area is the region of the brain associated with (spoken) language.

[see page 6, Applications]

• Information Retrieval Deciding relvence of a bunch of documents based on a search query, think Google.
• Information Extraction Recognise (Specific) information from text. eg. Foo IS A Dog.
• Text Categorisation Put text into discrete categories, eg. Email into Spam
• Summarisation Extract essential information from (one or more) text articles.
• Natural Language Generation Generate natural language text from an abstract, structured, representation. Eg. generate a manual in multiple languages from a single abstract representation of the steps to be carried out.
• Machine Translation English to French and/or vice versa. (VERY Difficult).

Considerations

Text isn't simple [see page 4, unstructured] information. It has structure, headings, tables etc. These are clues to help determine the importance of terms.