Information processing apparatus of displaying text with semantic segments thereof marked and determining and displaying candidate operations matching user intent corresponding to the text, information processing method thereof, and computer-readable storage medium

Patent metadata

Field	Value
Publication number	US-11086512-B2
Application number	US-201815933803-A
Country	US
Kind code	B2
Filing date	Mar 23, 2018
Priority date	Mar 27, 2017
Publication date	Aug 10, 2021
Grant date	Aug 10, 2021

Abstract

An information processing apparatus has a touch controllable display for displaying information, and an information processing method includes detecting a specific touch controlled operation on information including text displayed on a display, obtaining the text in the information touched by the specific touch controlled operation as a text-of-interest according to the detected specific touch controlled operation, performing a first natural language understanding processing which is based on word segmentation and named entity recognition on the text-of-interest to obtain a result of the word segmentation and a result of the named entity recognition for the text-of-interest, performing a second natural language understanding processing which is based on semantic analysis on the text-of-interest to obtain structured semantics of the text-of-interest in unit of semantic segments, and displaying the text-of-interest on the display in a manner that the semantic segments are marked.

First claim

What is claimed is:

1. An information processing method for an information processing apparatus having a touch screen for displaying information, the information processing method comprising:
detecting a specific touch operation on information including a text displayed on the touch screen;
obtaining the text in the information as a text-of-interest according to the detected specific touch operation;
performing a first natural language understanding processing based on word segmentation and named entity recognition on the text-of-interest to obtain a result of the word segmentation and a result of the named entity recognition for the text-of-interest;
performing a second natural language understanding processing based on semantic analysis on the text-of-interest to obtain structured semantics of the text-of-interest in unit of semantic segment, wherein the structured semantics includes semantic segments and a class of each of the semantic segments;
generating a display interface and assigning one view control to each of the semantic segments; and
displaying the semantic segments in a separated manner, wherein the semantic segments are arranged in separately-displayed view controls on the display interface.

2. The information processing method according to claim 1 , further comprising:
performing user intent recognition on the text-of-interest to obtain a user intent corresponding to the text-of-interest;
determining one or more candidate operations matching the user intent based on a predefined dictionary, and
displaying the determined candidate operations in an operable manner on the same display interface that presents the semantic segments in the separately-displayed view controls.

3. The information processing method according to claim 2 , wherein a semantic segment related to the obtained respective user intent among the semantic segments of the text-of-interest is further stored as a key semantic segment in association with the user intent, and the information processing method further comprising:
detecting a selection to the displayed candidate operations by the user; and
evoking the candidate operation selected by the user by using a key semantic segment which is associated with the user intent matching the respective selected candidate operation as a semantic segment for importing into the candidate operation to be evoked.

4. The information processing method according to claim 3 , further comprising:
determining whether a selection to the displayed semantic segments by the user is detected; and
using the semantic segment selected by the user as the text-of-interest so as to apply the second natural language understanding processing to the semantic segment selected by the user upon determining that the selection to the displayed semantic segments by the user is detected.

5. The information processing method according to claim 3 , wherein evoking the candidate operation further comprises:
generating a calling instruction for calling the candidate operation based on the structured semantics and the user intent, the calling instruction including evoking instruction and instruction argument;
extracting the key semantic segment as instruction argument according to the calling instruction triggered by the detected selection by the user; and
importing the key semantic segment into the candidate operation selected by the user through the instruction argument and evoking the candidate operation selected by the user by using the evoking instruction.

6. The information processing method according to claim 2 , wherein
the user intent recognition is further performed on each of the semantic segments of the text-of-interest to obtain the user intent corresponding to each of the semantic segments that can reflect the user intent.

7. The information processing method according to claim 2 , further comprising:
displaying the text-of-interest in a manner that a key semantic segment related to the user intent among the semantic segments of the text-of-interest is marked on type basis, wherein different types of key semantic segments is marked differently; and
wherein the candidate operation is a candidate application program or a candidate website link that is capable of matching the user intent.

8. The information processing method according to claim 2 , further comprising:
sorting the determined candidate operations to obtain a ranking of the determined candidate operations, and
displaying the selected candidate operations in accordance with the ranking.

9. The information processing method according to claim 1 , wherein the text-of-interest is displayed in a manner of word segmentation with a semantic segment mark.

10. An information processing apparatus having a touch screen for displaying information, the information processing apparatus comprising:
a detecting unit for detecting a specific touch operation on information including a text displayed on the touch screen;
a text-of-interest obtaining unit for obtaining the text in the information as a text-of-interest according to the detected specific touch operation;
a first natural language understanding unit for performing a first natural language understanding processing based on word segmentation and named entity recognition on the text-of-interest to obtain a result of the word segmentation and a result of the named entity recognition for the text-of-interest;
a second natural language understanding unit for performing a second natural language understanding processing which is based on semantic analysis on the text-of-interest to obtain structured semantics of the text-of-interest in unit of semantic segment, wherein the structured semantics includes semantic segments and a class of each of the semantic segments; and
a display unit for generating a display interface and assigning one view control to each of the semantic segments, and displaying the semantic segments in a separated manner, wherein the semantic segments are arranged in separately-displayed view controls on the display interface,
wherein the detecting unit, the text-of-interest obtaining unit, the first natural language understanding unit, the second natural language understanding unit, and the display unit are implemented via a processor.

11. The information processing apparatus according to claim 10 , further comprising:
a user intent obtaining unit for performing user intent recognition on the text-of-interest to obtain a user intent corresponding to the text-of-interest; and
a candidate operation determining unit for determining one or more candidate operations matching the user intent based on a predefined dictionary,
wherein the display unit further displays the determined candidate operations in an operable manner on the same display interface that presents the semantic segments in the separately-displayed view controls, and
wherein the user intent obtaining unit and the candidate operation determining unit are implemented via the processor.

12. The information processing apparatus according to claim 11 , wherein
the user intent obtaining unit further stores a semantic segment related to the obtained respective user intent among the semantic segments of the text-of-interest as a key semantic segment in association with the user intent, and the information processing apparatus further comprising:
a user selection accepting unit for detecting a selection to the displayed candidate operations by the user; and
an operation evoking unit for evoking the candidate operation selected by the user by using a key semantic segment which is associated with the user intent matching the respective selected candidate operation as

Assignees

• Tencent Tech Shenzhen Co Ltd

Inventors

• Hou Xinglin
• Qi Chao
• Wang Zhuoran
• Ma Yuchi

Classifications

• G06F40/30 Primary

  Semantic analysis · CPC title

• H04M1/72436

  for text messaging, e.g. short messaging services [SMS] or e-mails · CPC title

• G06F3/0412

  Digitisers structurally integrated in a display · CPC title

• G06F3/0484

  for the control of specific functions or operations, e.g. selecting or manipulating an object, an image or a displayed text element, setting a parameter value or selecting a range · CPC title

• G06Q10/109

  Time management, e.g. calendars, reminders, meetings or time accounting · CPC title