Predictive Text

Last updated 4 months ago

Can you work out the missing or swapped two letter words from a list of sentences faster than your opponent?

Objective

This game has two different variations.

  • You are given a list of sentences and all of the two letter words have been removed.

  • You are given a list of sentences and some of the two letter words have been swapped with another two letter word. There is a fifty percent chance a two letter word will be swapped with another incorrect two letter word.

Both games however have the same objective that you must replace or add the correct two letter word before your opponent. You can submit an answer at any time but you are only allowed to submit one replacement or addition at a time. You are not allowed to submit the same answer as your opponent, so that means if you suspect your opponent has the correct answer you need to move on and look at another missing or swapped word.

Only at the end of the game are you given the correct answers. Answers are not case sensitive e.g. "It" is the same as "it". The player with the most correct answers is the winner. In the event both you and your opponent submit the same number of correct answers the game will be a draw.

Strategy

Understanding natural language is a complex area. There are many resources on the net to help with grammar and word prediction. Areas to look at are :

  • Natural Language Toolkit : NLTK is a leading platform for building Python programs to work with human language data.

  • Microsoft Cognitive Services : Automate a variety of standard, natural language processing tasks using state-of-the-art language modelling APIs.

  • Stanford CoreNLP : Stanford CoreNLP provides a set of human language technology tools. It can give the base forms of words, their parts of speech, whether they are names of companies, people, etc., normalize dates, times, and numeric quantities, mark up the structure of sentences in terms of phrases and syntactic dependencies, indicate which noun phrases refer to the same entities, indicate sentiment, extract particular or open-class relations between entity mentions, get the quotes people said, etc.