CN-121981117-A - Information identification method and related device

Abstract

The application provides an information identification method and a related device, and relates to the technical field of terminals. The method comprises the steps of firstly responding to triggering operation of an arbitrary gate interaction entrance to obtain first information, then carrying out entity identification of a first category on the first information through an entity identification model to obtain an entity belonging to the first category in the first information, training the entity identification model based on a first training sample and a second training sample, wherein the first training sample comprises the entity of the first category, the entity of the first category is provided with a first label, the second training sample comprises the entity of the second category, the entity of the second category is provided with a second label, the first label and the second label belong to the first category, then determining application service based on the entity of the first information, and finally displaying indication information of application providing the application service. Therefore, the accuracy of the entity recognition model in recognizing the unseen information can be improved, and further application services which are more in line with the actual requirements of the users can be recommended for the users.

Inventors

• WANG YURAN
• ZHOU YUHAO

Assignees

• 荣耀终端股份有限公司
• 复旦大学

Dates

Publication Date

20260505

Application Date

20241031

Claims (14)

1. 1. An information identification method, comprising: responding to the triggering operation of any door interaction entrance to acquire first information; performing entity identification of a first category on the first information through an entity identification model to obtain an entity of the first information; The entity identification model is obtained by training based on a training sample set, the training sample set comprises a first training sample and a second training sample, the first training sample comprises entities of the first category, the entities of the first category are provided with first labels, the second training sample comprises entities of the second category, the entities of the second category are provided with second labels, and the first labels and the second labels belong to the first category; Determining an application service based on the entity of the first information; Displaying indication information of an application providing the application service.
2. 2. The method as recited in claim 1, further comprising: obtaining a first entity based on the related information of the user, wherein the first entity belongs to the first category; Filtering the entity of the first information based on the first entity; the determining, by the entity based on the first information, an application service includes: The application service is determined based on the entity of the filtered first information.
3. 3. The method of claim 2, wherein the entity of the first information comprises a second entity, and wherein the filtering the entity of the first information based on the first entity comprises: performing similarity matching based on the first entity and the second entity; And filtering out the second entity of which the matching is unsuccessful from the second entities.
4. 4. The method of claim 3, wherein the performing similarity matching based on the first entity and the second entity comprises: If the ratio of the number of repeated characters between the characters of the first entity and the characters of the second entity to the number of characters of the first entity is greater than or equal to a second threshold value, determining that the second entity is successfully matched; and if the ratio of the number of repeated characters between the characters of the first entity and the characters of the second entity to the number of characters of the first entity is smaller than the second threshold value, determining that the second entity is not successfully matched.
5. 5. The method of claim 2, wherein the first category comprises a location, wherein the obtaining the first entity based on the user-related information comprises: Obtaining the first entity based on user location information; And/or the number of the groups of groups, And obtaining the first entity based on the user service information.
6. 6. The method of claim 5, wherein the obtaining the first entity based on user location information comprises: inquiring a first interest point within a first preset distance of the user position information; Generating a place abbreviation of the first interest point as the first entity through a place abbreviation generation model; And/or the number of the groups of groups, And inquiring a first place abbreviation within a second preset distance of the user position information from a first database as the first entity, wherein the first database stores a plurality of place abbreviations and position information corresponding to the place abbreviations respectively.
7. 7. The method according to any of claims 1-6, wherein the first tag is derived based on binary encoding and the second tag is derived based on smooth encoding.
8. 8. The method of claim 7, wherein the first class of entities has a first class tag and the second class of entities has a second class tag, the first class tag being identical to the second class tag, the first tag being obtained by binary encoding the first class tag and the second tag being obtained by smooth encoding the second class tag.
9. 9. The method of claim 7, wherein the trained entity class probability distribution output by the entity recognition model is obtained by correcting a first probability among initial entity class probabilities based on a third threshold, the first probability including a probability corresponding to the first class.
10. 10. The method of any of claims 1-6, wherein the first tag and the second tag are each derived based on binary encoding.
11. 11. The method of claim 10, wherein the first class of entities has a third class label and the second class of entities has a fourth class label, the third class label being different from the fourth class label, the first label being obtained by binary encoding the third class label and the second label being obtained by binary encoding the fourth class label.
12. 12. An electronic device comprising a memory and a processor; The memory being coupled to the processor, the memory being for storing computer program code comprising computer instructions that one or more of the processors call to cause the electronic device to perform the information identification method of any of claims 1-11.
13. 13. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements the information identification method according to any of claims 1-11.
14. 14. A computer program product comprising computer program code which, when executed by an electronic device, implements the steps of the information identification method of any of claims 1-11.

Description

Information identification method and related device Technical Field The present application relates to the field of terminal technologies, and in particular, to an information identification method and a related device. Background With the continuous development of terminal technology, electronic devices can recognize various forms of information such as text, images, voice, or video, for example, electronic devices can recognize text from various forms of information including text. However, there may occur a problem in that the information is erroneously recognized, taking a text including a place as an example, a place for a place is abbreviated as a place, etc., the electronic device may not recognize it as a place, but as a text of another category. Further, in some scenarios, an application service meeting the requirements of the user cannot be provided for the user, for example, in any door scenario, a place included in the information is identified as text of other types, and the electronic device may not be capable of recommending an application icon (such as an icon of a map navigation application) meeting the actual requirements of the user, thereby affecting the use experience of the user. Disclosure of Invention In order to solve the problems, the application provides an information identification method and a related device, which aim to improve the accuracy of information identification, so that electronic equipment can recommend application icons meeting the actual demands of users under any scenes such as doors and the like, and the use experience of the users is improved. In a first aspect, the present application provides an information identifying method, which may be applied to an electronic device, for example, a mobile phone, a tablet computer, a notebook computer, or the like. In the method, the electronic device responds to a triggering operation of any portal interaction portal to acquire first information, the triggering operation of any portal interaction portal can comprise a long-press operation of a user on the first information and an operation of dragging the first information to any side (such as the left side or the right side) of a screen, wherein the distance between the first information and the screen is smaller than or equal to a preset distance, the electronic device can acquire the first information to identify the first information, the first information can be a text, an image, voice or video and the like, then the electronic device can input the first information into an entity identification model, the entity identification model is used for identifying an entity of the first category in the first information, the entity identification model can output the entity of the first category in the first information, the first category is a place, the entity of the first information is an entity of the place, the electronic device can determine an application service based on the entity of the first information, the entity is an entity of the place, the electronic device can confirm that the electronic device can display an application name, the electronic device can provide an application map or the like based on the fact that the place information confirms the application name, the electronic device can provide the application service, can display the application name or the application map, and the like. The entity recognition model can be obtained by training based on a training sample set, the training sample set can comprise a first training sample and a second training sample, the first training sample comprises a first class of entities, the second training sample comprises a second class of entities, the first class is a place, the first training sample comprises a place entity, the second training sample can comprise other classes of entities, such as common words in other classes of entities, for example, a name entity, a movie entity and the like, the common words in other classes of entities can be noun words, such as football, a table and the like, the common words can also be new words, and words composed of a plurality of high-frequency single words in the noun words, such as Baxin, tree days and the like. The first class of entities has a first tag and the second class of entities has a second tag, both the first tag and the second tag belonging to the first class, the first tag of the first class being obtained by encoding a B-tag in the BIO-tag, the second tag of the second class being obtained by encoding a B-tag in the BIO-tag, both the first tag and the second tag belonging to a tag of the location entity, and the second tag of the second class being obtained by encoding a B-tag in the BIO-tag, the second tag of the second class being obtained by encoding a B (replayed) -tag in the BIO-tag, both the first tag and the second tag belonging to a tag of the location entity. Therefore, the recall rate of the place recognition model trained based on th