Data Input System Using N-gram Ranking for Text Item Context
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face difficulties in understanding new text items such as emojis, hashtags, and rare words in data input systems, requiring manual navigation to external dictionaries, which is cumbersome and disrupts the workflow.
Innovation Solution
A computer-implemented data input system that uses an n-gram language model to identify and rank examples of use for a given text item, allowing users to request and view explanations without leaving the current application, utilizing both static n-gram and dynamic models to provide context-specific examples.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If users manually navigate to external dictionary websites to understand unfamiliar text items, then they can find meanings of text items, but the workflow is disrupted and time is lost
Solution Approach 1:
The patent combines the dictionary lookup function directly within the messaging application interface. When a user encounters an unfamiliar text item (emoji, hashtag, word), they can tap it to see examples of use without leaving the current application. This merges the previously separate functions of messaging and dictionary lookup into a single integrated system, eliminating the need to navigate to external websites and preserving the user's workflow.
2Loss of information
If users manually copy and paste text items into external query boxes, then they can search for meanings, but the operation becomes cumbersome
Solution Approach 1:
The system automatically detects when a user taps on an unfamiliar text item and automatically retrieves and displays examples of use for that item. The user does not need to manually copy, paste, or type the text item into a search box - the system services the user's information need automatically through a simple tap gesture, greatly simplifying the operation.
3Productivity
If predictive technology provides candidate next data items, then data input efficiency is improved, but users may encounter unfamiliar text items they do not understand
Solution Approach 1:
The system provides immediate feedback when a user encounters an unfamiliar text item in the predictive candidate list. By tapping on the unfamiliar item, the user receives instant examples of use that help them understand the meaning and decide whether to select it. This feedback loop allows users to efficiently evaluate predictive suggestions without breaking their data input flow, maintaining high productivity while improving comprehension.
Data Source
AI summary
A computer-implemented example generator is described which has a memory storing a text item, examples of use of the text item having been requested. A processor searches at least one n-gram language model to identify n-grams containing the text item. The processor is configured to rank the identified n-grams on the basis of a position of the text item in the identified n-grams; and a user interface presents at least some of the identified n-grams to a user taking into account the ranking.