Emoji for Text Predictions

• US 20150100537A1
• Filed: 10/03/2013
• Published: 04/09/2015
• Est. Priority Date: 10/03/2013
• Status: Abandoned Application

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Grieves, Jason A., Almog, Itai, Badger, Eric Norman, Cook, James H., Garcia Fierro, Manuel
• Application Number:
  US14/045,461
• Publication Number:
  US 20150100537A1
• Time in Patent Office: Days
• US Class Current: 706/52
• CPC Class Codes: G06F 3/0236 using selection techniques to select from displayed items
• CPC Class Codes: G06F 3/0237 using prediction or retrieval techniques
• CPC Class Codes: G06F 40/274 Converting codes to words; Guess-ahead of partial word inputs
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 5/025 Extracting rules from data
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06N 5/048 Fuzzy inferencing
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

Techniques to employ emoji for text predictions are described herein. In one or more implementations, entry of characters is detected during interaction with a device. Prediction candidates corresponding to the detected characters are generated according to a language model that is configured to consider emoji along with words and phrases. The language model may make use of a mapping table that maps a plurality of emoji to corresponding words. The mapping table enables a text prediction engine to offer the emoji as alternatives for matching words. In addition or alternatively, the text prediction engine may be configured to analyze emoji as words within the model and generate probabilities and candidate rankings for predictions that include both emoji and words. User-specific emoji use may also be learned by monitoring a user'"'"'s typing activity to adapt predictions to the user'"'"'s particular usage of emoji.

Claims

1. A method, comprising:

detecting entry of characters during interaction with a device;

generating one or more prediction candidates including one or more predicted emoji corresponding to the detected characters according to a language model; and

employing the one or more prediction candidates to facilitate further character entry for the interaction with the device.

2. A method as recited in claim 1, wherein the language model is configured to adapt predictions made by a text prediction engine to typing styles of users on an individual basis including user-specific emoji usage.

3. A method as recited in claim 1, wherein the language model is designed to make use of multiple language model dictionaries as sources of words, emoji, and corresponding scoring data, the scoring data tuned based on collection and analysis of word and emoji usage data for collective typing activities of a population of users.

4. A method as recited in claim 1, wherein generating the one or more prediction candidates comprises computing a weighted combination of scoring data associated with words and emoji contained in multiple dictionaries associated with the language model to compute scores for the prediction candidates.

5. A method as recited in claim 4, wherein generating the one or more text prediction candidates further comprises ranking the prediction candidates including words and emoji one to another based on the computed scores.

6. A method as recited in claim 1, further comprising collecting data regarding usage of emoji on a user-specific basis to create a user-specific dictionary for the language model that reflects usage of emoji.

7. A method as recited in claim 1, wherein generating the one or more prediction candidates comprises identifying one or more emoji as prediction candidates that correspond to predicted words based upon a mapping table for the language model that directly maps emoji to words.

8. A method as recited in claim 1, further comprising enabling switching between emoji and words during interaction with the device based upon a mapping table associated with the language model that directly maps emoji to words including:

switching between a particular word and a corresponding emoji responsive to a selection of the particular word to cause the switching; and

switching between a particular emoji and a corresponding word responsive to a selection of the particular emoji to cause the switching.

9. A method as recited in claim 1, wherein employing the one or more prediction candidates comprises presenting representations of one or more prediction candidates via a user interface of the device for selection by a user to automatically insert a selected candidate to modify the detected characters by replacement or insertion after the detected characters.

10. A method as recited in claim 1, wherein employing the one or more prediction candidates comprises presenting representations of one or more prediction candidates for selection by a user via a prediction bar exposed in connection with an on-screen keyboard of the device.

11. A method as recited in claim 10, wherein the prediction candidates that are emoji and prediction candidates that are words are interspersed in the prediction bar.

12. A method as recited in claim 10, wherein the prediction bar is configured to present prediction candidates that are emoji and prediction candidates that are words as separate groups of prediction candidates.

13. A method as recited in claim 1, wherein employing the one or more prediction candidates comprises exposing a ranked list of prediction candidates for selection by a user to modify the detected characters, the ranked list of prediction candidates including at least one predicted emoji.

14. A method as recited in claim 14, further comprising:

responsive to interaction with the at least one predicted emoji configured to access an emoji picker, displaying the emoji picker configured to present and enable selection of a plurality of emoji options that relate to the at least one predicted emoji.

15. One or more computer-readable storage media storing instructions that, when executed by a computing device, cause the computing device to perform operations comprising:

identifying one or more dictionaries to use as sources for predictions based on one or more detected characters;

ranking emoji along with words one to another as prediction candidates for the detected characters using scoring data contained in the one or more dictionaries;

selecting one or more top ranking emoji and words according to the ranking as prediction candidates for the detected characters; and

utilizing selected emoji along with selected words to facilitate character entry.

16. One or more computer-readable storage media as recited in claim 15, wherein the multiple dictionaries comprise a general population dictionary representative of common usage across a community of users and at least one other dictionary generated dynamically based on input of words and emoji by a particular user of the computing device to reflect the particular user'"'"'s individual typing style.

17. One or more computer-readable storage media as recited in claim 15, wherein utilizing the selected emoji along with the words to facilitate character entry comprises:

representing multiple different emoji that are determined as top ranking prediction candidates along with words that are determined as top ranking prediction candidates via a user interface instrumentality configured to enable a selection from among the prediction candidates to modify the one or more detected characters.

and

enabling switching between words of the detected one or more characters and emoji that are directly mapped to the words responsive to selection of the words to cause the switching.

18. A mobile computing device, comprising:

a processing system; and

one or more computer-readable media storing instructions that, when executed by the processing system, implement a text prediction engine operable to;

generate one or more prediction candidates for characters detected in an interaction scenario according to one or more dictionaries of a language model that support emoji, the prediction candidates that are generated including emoji and words predicted using the language model;

exposing the prediction candidates that are generated to enable selection from among the prediction candidates to modify the one or more detected characters.

19. A computing device as recited in claim 18, wherein exposing the prediction candidates comprises exposing at least one predicted word in connection with a corresponding emoji that is determined based upon a mapping table of the language model configured to directly map a plurality of words to corresponding emoji.

20. A computing device as recited in claim 18, wherein:

the one or more dictionaries are configured to include conditional usage probabilities for language-specific usage of emoji for in relation to the interaction scenario and; and

the one or more prediction candidates are generated and ranked one to another based at least in part upon the conditional usage probabilities for language-specific usage of emoji.