CN114202087A

CN114202087A - Information processing method and computing device

Info

Publication number: CN114202087A
Application number: CN202010988002.1A
Authority: CN
Inventors: 张淼; 蒋保生
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2020-09-18
Filing date: 2020-09-18
Publication date: 2022-03-18

Abstract

The embodiment of the application provides an information processing method and computing equipment, wherein the information processing method comprises the following steps: acquiring an object to be detected provided by a target user; inputting the object to be detected into a grade prediction model to obtain an initial passing grade of the object to be detected; performing grade optimization processing on the initial passing grade by using a grade optimization model to obtain a target passing grade of the object to be detected; and outputting the target passing grade for the target user. According to the embodiment of the application, the object is effectively submitted by accurately predicting the passing grade.

Description

Information processing method and computing device

Technical Field

The present application relates to the field of electronic technologies, and in particular, to an information processing method and a computing device.

Background

The trademark is a mark for distinguishing the commodity or service of the operator from the commodity or service of another operator, and may specifically include a combination of elements such as characters, graphics, numbers, sounds, three-dimensional symbols, and colors. Currently, prior to use, trademarks require the submission of a trademark application to the trademark office, which reviews trademarks submitted by trademark applicants. However, when the trademark applicant submits the trademark, whether the applied trademark can be checked or not cannot be known in time, and some invalid trademarks are easily submitted, so that the success rate of trademark application is reduced.

Disclosure of Invention

In view of this, embodiments of the present application provide an information processing method and a computing device, so as to solve the technical problem of low success rate of applying a trademark due to invalid submission of the trademark in the prior art.

In a first aspect, an embodiment of the present application provides an information processing method, including:

acquiring an object to be detected provided by a target user;

inputting the object to be detected into a grade prediction model to obtain an initial passing grade of the object to be detected;

performing grade optimization processing on the initial passing grade by using a grade optimization model to obtain a target passing grade of the object to be detected;

and outputting the target passing grade for the target user.

In a second aspect, an embodiment of the present application provides an information processing method, including:

acquiring a trademark to be detected provided by a target user;

inputting the trademark to be detected into a grade prediction model to obtain the initial passing grade of the trademark to be detected;

performing grade optimization processing on the initial passing grade by using a grade optimization model to obtain a target passing grade of the trademark to be detected;

and outputting the target passing grade for the target user.

In a third aspect, an embodiment of the present application provides a computing device, including: a storage component and a processing component; the storage component is used for storing one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component;

the processing component is to:

acquiring an object to be detected provided by a target user; inputting the object to be detected into a grade prediction model to obtain an initial passing grade of the object to be detected; performing grade optimization processing on the initial passing grade by using a grade optimization model to obtain a target passing grade of the object to be detected; and outputting the target passing grade for the target user.

In a fourth aspect, an embodiment of the present application provides a computing device, including a storage component and a processing component; the storage component is used for storing one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component;

the processing component is to:

acquiring a trademark to be detected provided by a target user; inputting the trademark to be detected into a grade prediction model to obtain the initial passing grade of the trademark to be detected; performing grade optimization processing on the initial passing grade by using a grade optimization model to obtain a target passing grade of the trademark to be detected; and outputting the target passing grade for the target user.

According to the method and the device, the object to be detected provided by the target user is obtained, so that the object to be detected can be input into the grade prediction model to obtain the initial passing grade of the object to be detected. And carrying out preliminary prediction on the passing grade of the object to be detected through a grade prediction model. And then, performing grade optimization processing on the initial passing grade by using a grade optimization model to obtain the target passing grade of the object to be detected. And performing further grade optimization on the initial grade by using a grade optimization model so as to improve the precision of the obtained target passing grade. Therefore, when the target passing grade is output for the target user, effective passing grade prompt can be provided for the user, and passing efficiency is improved.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative efforts.

Fig. 1 is a flowchart of an embodiment of an information processing method according to an embodiment of the present application;

fig. 2 is a flowchart of another embodiment of an information processing method according to an embodiment of the present application;

fig. 3 is a flowchart of another embodiment of an information processing method according to an embodiment of the present application;

fig. 4 is a flowchart of another embodiment of an information processing method according to an embodiment of the present application;

fig. 5 is a diagram illustrating a display example of prompt information for multiple categories according to an embodiment of the present application;

fig. 6 is a diagram illustrating a display example of prompt information for multiple categories according to an embodiment of the present application;

fig. 7 is a diagram illustrating a display example of prompt information for multiple categories according to an embodiment of the present application;

fig. 8 is a flowchart of another embodiment of an information processing method according to an embodiment of the present application;

fig. 9 is a flowchart of another embodiment of an information processing method according to an embodiment of the present application;

fig. 10 is a diagram illustrating an example of an information processing method according to an embodiment of the present application;

FIG. 11 is a block diagram illustrating an embodiment of a computing device, according to an embodiment of the present disclosure;

fig. 12 is a schematic structural diagram of an embodiment of a computing device according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The terminology used in the embodiments of the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in the examples of this application and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise, and "a" and "an" typically include at least two, but do not exclude the presence of at least one.

It should be understood that the term "and/or" as used herein is merely one type of association that describes an associated object, meaning that three relationships may exist, e.g., a and/or B may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" herein generally indicates that the former and latter related objects are in an "or" relationship.

The words "if," "if," as used herein may be interpreted as "at … …" or "at … …" or "in response to a determination" or "in response to a recognition," depending on the context. Similarly, the phrases "if determined" or "if identified (a stated condition or event)" may be interpreted as "when determined" or "in response to a determination" or "when identified (a stated condition or event)" or "in response to an identification (a stated condition or event)", depending on the context.

It is also noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a good or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such good or system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a commodity or system that includes the element.

The technical scheme of the embodiment of the application can be applied to effective judgment of trademarks. Aiming at the object to be detected, particularly when the object to be detected is a trademark, the passing grade of the object to be detected can be accurately predicted by combining a grade prediction model and a grade optimization model, so that the user passing rate is improved.

In the prior art, the trademark applicant may submit a trademark to the trademark office. The trademark office examines whether the trademark filed by the applicant can pass. But applicants submit trademarks that are not known to be acceptable for examination. Generally, in order to know the passability of the trademark, the comparison trademark obtained by detection is simply judged by simply searching the trademark so as to cause that the possibility of passing the application of the trademark by the applicant is unknown, so that invalid trademarks are easily submitted and invalid applications are generated.

In the embodiment of the application, the object to be detected of the target user can be obtained, and then the object to be detected can be input into the grade prediction model to obtain the initial passing grade of the object to be detected. At this time, the passing grade of the object to be detected is preliminarily predicted by using the grade prediction model. And then, carrying out grade optimization processing on the initial passing grade by using a grade optimization model to obtain the target passing grade of the object to be detected. At the moment, the grade optimization model is utilized to perform further grade optimization on the initially predicted passing grade, and the target passing grade accurately measured on the object to be detected is obtained, so that when the target user outputs the target passing grade, the passing possibility of the object to be detected of the target user can be effectively prompted through the accurate target passing grade, and the generation of invalid objects is avoided.

The embodiments of the present application will be described in detail below with reference to the accompanying drawings.

As shown in fig. 1, which is a flowchart of an embodiment of an information processing method provided in an embodiment of the present application, the method may include the following steps:

101: and acquiring the object to be detected provided by the target user.

The information processing method provided by the embodiment of the application can be applied to computing equipment or a server. The computing device may include, for example: the embodiment of the application does not limit the specific type of the computing equipment. The server may specifically include: the embodiment of the application does not limit the specific types of the servers too much.

When the technical scheme provided by the application is applied to the computing equipment, the object to be detected can be acquired by the computing equipment. When the technical scheme provided by the application is applied to the server, the object to be processed can be collected by the user side used by the target user and sent to the server.

The object to be detected can be provided for a target user. In practical applications, the object to be detected may be a trademark that needs to be confirmed with possibility. In addition, the object to be detected may be an image, a character, a video, or the like, which generally needs to be confirmed with possibility.

In the passing possibility judgment scenario of a trademark, the target user may be a trademark applicant.

102: and inputting the object to be detected into a grade prediction model to obtain the initial passing grade of the object to be detected.

Alternatively, the class prediction model may be obtained by training in advance. The model parameters of the grade prediction model used in step 102 are known, and at this time, the object to be detected is input into the grade prediction model with known parameters, so as to obtain the initial passing grade of the object to be detected.

The level prediction model may be a deep neural network model, and the level prediction model may be obtained by performing parameter training on the level prediction model by using a training object and a reference result corresponding to the training object. The grade prediction model in the embodiment of the present application may be a deep neural network model, and the specific type of the deep neural network model in the embodiment of the present application is not limited too much.

The initial pass level may be a pass level obtained by preliminarily predicting, by the level prediction model, a likelihood of passing of the object to be detected.

The level prediction model may perform level prediction on an input object to be detected, where at least one passing level may be preset in the level prediction model, and for example, the at least one passing level may include: low level, lower level, medium level, higher level, high level, and the like.

The initial passing level may be any one of at least one preset passing level in the level prediction model, for example, when the level prediction model predicts a similar object whose meaning has a similarity of 90% to the object to be detected, the initial passing level may be determined to be a low passing level.

103: and performing grade optimization processing on the initial passing grade by using a grade optimization model to obtain the target passing grade of the object to be processed.

The rank optimization model may perform a rank optimization process on the initial pass rank. In practical applications, whether the trademark is approved or not is influenced by the examination rule in addition to the similarity, for example, the same applicant can apply a second application on the basis of the existing trademark, and the registered trademark of the second application is more likely to pass. And carrying out grade optimization processing on the initial passing grade through a grade optimization model, wherein the obtained target passing grade has higher matching degree with the actual trademark examination and passing possibility, and the accuracy of the target passing grade is improved.

Alternatively, in a trademark review scenario, the rating optimization model may be determined by some review rules for trademarks. Of course, in order to implement the automated examination of the examination rule, the level optimization model may be designed as some examination flow processing steps, and reference may be specifically made to the description of the subsequent embodiments.

104: and outputting the target passing grade for the target user.

When the embodiment of the application is applied to the computing device, the computing device can directly output the target passing grade for the target user through the display component.

When the embodiment of the application is applied to the server, the server can provide the target passing grade for the user side corresponding to the target user, and the display component of the user side outputs the target passing grade for the target user.

In the embodiment of the application, the object to be detected of the target user can be obtained, and then the object to be detected can be input into the grade prediction model to obtain the initial passing grade of the object to be detected. At this time, the passing grade of the object to be detected is preliminarily predicted by using the grade prediction model. And then, carrying out grade optimization processing on the initial passing grade by using a grade optimization model to obtain the target passing grade of the object to be detected. At the moment, the grade optimization model is utilized to perform further grade optimization on the initially predicted passing grade, and the target passing grade accurately measured on the object to be detected is obtained, so that when the target user outputs the target passing grade, the passing possibility of the object to be detected of the target user can be effectively prompted through the accurate target passing grade, and the generation of invalid objects is avoided. Especially, when the object to be detected is the trademark, the success rate of trademark registration can be improved.

As shown in fig. 2, a schematic structural diagram of another embodiment of an information processing method provided in the embodiment of the present application is shown, where the method may include the following steps:

201: and acquiring the object to be detected provided by the target user.

202: and inputting the object to be detected into the similar object detection submodel to obtain at least one similar object matched with the object to be detected and the similarity between the at least one similar object and the object to be detected respectively.

203: and based on the similarity corresponding to at least one similar object, carrying out grading processing according to the grading sub-model to obtain the initial passing grade of the object to be detected.

Wherein, the grade optimization model can include: a similar object detection submodel and a hierarchical submodel.

The similar object detection sub-model may detect a similar object for the object to be detected, and in practical applications, the similar object detection model may be a deep neural network model, and may search for a similar object similar to the object to be detected from the database, and determine a similarity between the similar object and the object to be detected.

The similar object may be an object similar to the object to be detected. Taking the example that the object to be detected is "Riverle", the similar object obtained by detecting the object to be detected by the similar object detection submodel can be "Riverest", "Reverle", "Mi le", and the like, at this time, the similarity between the object to be detected and "Riverle" can be "0.965", the similarity between the object to be detected and "Riverle" is 0.574, the similarity between the object to be detected and "Mi le" is 0.965, and the similarity between the object to be detected and "Mi le" is 0.965.

The passing grade of the object to be detected is inversely related to the similarity of the corresponding at least one similar object. If the similarity between the object to be detected and the similar object is higher, the two objects are more similar, and the possibility that the object to be detected passes the examination is lower at the moment; if the similarity between the object to be detected and the similar object is higher, the two objects are more dissimilar, and the possibility that the object to be detected passes the examination is high. That is, the similarity corresponding to each of the at least one similar object is inversely proportional to at least one passing level in the level division submodel. The higher the similarity, the lower the category level of the passing level, the lower the similarity, the higher the category level of the passing level. In some embodiments, the ranking sub-model may preset at least one passing rank, for example, the at least one passing rank may include: low level, lower level, medium level, higher level, high level, and the like. Based on the similarity corresponding to each of the at least one similar object, performing a ranking process according to the ranking submodel, and obtaining an initial passing rank of the object to be detected may include: and determining the passing grade matched with the object to be detected in at least one passing grade as the initial passing grade according to the similarity corresponding to the at least one similar object respectively.

204: and performing grade optimization processing on the initial passing grade by using a grade optimization model to obtain the target passing grade of the object to be detected.

205: and outputting the target passing grade for the target user.

In the embodiment of the application, after the object to be detected provided by the target user is obtained, the object to be detected may be input into the similar object detection submodel, so as to obtain at least one similar object matched with the object to be detected and the similarity between the at least one similar object and the object to be detected. Whether an object with higher similarity to the object to be detected exists or not can be judged by detecting the similar objects, and then, grading processing is carried out according to the similarity between at least one searched similar object and the object to be detected so as to obtain the initial passing grade of the object to be detected. Effective preliminary grade division can be realized through grade division of the similarity detection result, the initial passing grade is obtained, and the accuracy of the initial passing grade is improved. And then, carrying out grade optimization processing on the initial passing grade by using a grade optimization model to obtain the target passing grade of the object to be detected. At the moment, the grade optimization model is utilized to perform further grade optimization on the initially predicted passing grade, and the target passing grade accurately measured on the object to be detected is obtained, so that when the target user outputs the target passing grade, the passing possibility of the object to be detected of the target user can be effectively prompted through the accurate target passing grade, and the generation of invalid objects is avoided. Especially, when the object to be detected is the trademark, the success rate of trademark registration can be improved.

As an example, in order to clearly show the high and low of the passing possibility, different passing levels may be set. The grade division submodel may be provided with at least one passing grade, and the input object to be detected is matched with any one of the at least one passing grade, that is, the actual grade of the initial passing grade is obtained.

Based on the similarity corresponding to each of the at least one similar object, performing a ranking process according to the ranking submodel, and obtaining an initial passing rank of the object to be detected may include:

acquiring at least one passing grade corresponding to the grade division submodels respectively and a similarity range corresponding to the at least one passing grade respectively;

determining a target similarity range in at least one similarity range based on the similarity corresponding to at least one similar object respectively;

and determining the passing grade corresponding to the target similarity range as an initial passing grade.

Wherein the category level of at least one pass level is inversely related to the size of the similarity range. The higher the rank is, the smaller the similarity range of the passage rank is, and the lower the rank is, the larger the similarity range of the passage rank is.

For example, when at least one passing rank is a low rank, a lower rank, a middle rank, a higher rank, and a high rank, the respective corresponding similarity ranges may be, for example: low grade: the similarity range is more than 90%; lower grade: the similarity range is 80% -90%; medium grade: the similarity range is 60% -80%; and (3) high grade: the similarity range is 30-60%; and high grade: the similarity range is less than 30%.

The pass level corresponds to a similarity range, the initial pass level can be quickly determined through similarity range matching, and the effective speed of obtaining the pass level is improved.

As a possible implementation manner, determining the target similarity range in the at least one similarity range based on the similarity corresponding to the at least one similar object respectively may include:

determining the maximum similarity among the similarities corresponding to the at least one similar object respectively;

and determining a target similarity range corresponding to the maximum similarity in at least one similarity range.

Determining the maximum similarity among the similarities corresponding to the at least one similar object may include: and arranging the similarity corresponding to at least one similar object according to the sequence from big to small, wherein the obtained first similarity is the maximum similarity. Determining the target similarity range corresponding to the maximum similarity in the at least one similarity range may specifically include: and searching a target similarity range in which the maximum similarity is located in at least one similarity range, namely determining the similarity range in which the maximum similarity is located.

The target similarity range can be quickly determined through the maximum similarity.

As another possible implementation manner, determining the target similarity range in the at least one similarity range based on the similarity corresponding to the at least one similar object respectively may include:

determining similarity ranges to which the similarities corresponding to the at least one similar object respectively belong according to the similarity ranges corresponding to the at least one passing grade respectively so as to obtain the similarities corresponding to the at least one similarity range respectively;

counting the number of the similarity in any similarity range to obtain the number of similar objects in the similarity range so as to obtain the number of similar objects corresponding to at least one similarity range;

determining at least one candidate similarity range with the number of similar objects larger than 1;

and acquiring a target similarity range with the highest range in the at least one candidate similarity range.

According to the similarity ranges respectively corresponding to the at least one passing grade, determining the similarity ranges to which the similarities respectively corresponding to the at least one similar object belong, counting the number of the similarities in any one similarity range, obtaining the number of the similar objects in the similarity range, so as to obtain the number of the similar objects respectively corresponding to the at least one similarity range, namely, dividing the similarities belonging to the same similarity range into the same group, and counting the number of the similarities of each group. After obtaining the number of similar objects corresponding to at least one similarity range respectively, the similarity range with the number of similar objects larger than 1 may be used as the candidate similarity range.

After determining at least one candidate similarity range with the number of similar objects greater than 1, in some embodiments, after determining the similarity range to which the similarity corresponding to at least one similar object respectively belongs according to the similarity range corresponding to at least one passing level respectively to obtain the similarity corresponding to at least one similarity range respectively, the similar objects corresponding to the similarity in each similarity range may be subjected to invalid judgment, and the similarity corresponding to the similar object with invalid judgment is deleted from the similarity range to obtain the valid similarity corresponding to at least one similarity range respectively. At this time, the number of valid similarities in any similarity range may be counted, and the number of similar objects in the similarity range may be obtained, so as to obtain the number of similar objects corresponding to at least one similarity range respectively.

In an actual trademark review scenario, the rating optimization model may be determined by some review rules for trademarks. Of course, to enable automated review of review rules, the hierarchical optimization model may be designed as a number of review flow process steps.

As an embodiment, performing level optimization processing on the initial pass level by using a level optimization model, and obtaining a target pass level of the object to be detected may include:

determining a first category corresponding to the object to be detected at the designated category level;

determining second categories corresponding to the at least one similar object at the specified category level respectively;

judging whether a target category which is the same as the first category exists in second categories respectively corresponding to at least one similar object;

if so, determining a target pass level based on the initial pass level;

and if the target passing level does not exist, determining that the target passing level is the highest passing level in the at least one passing level corresponding to the grade division submodel.

Wherein the specified category level may be a subdirectory level under the first category level. The first category level is a highest category level under a category classification standard, and the category level of the specified category level is lower than the highest category level. Determining the first category corresponding to the object to be detected at the designated category level may include: and determining the first category corresponding to the object to be detected at the specified category level based on the third category of the object to be detected at the first category level.

The first category corresponding to the object to be detected at the designated category level may be the first category selected for the object to be detected at the designated category level by the target user.

Determining the second categories to which the at least one similar object respectively corresponds at the specified category level may include: and determining second categories corresponding to the at least one similar object respectively based on fourth categories of the at least one similar object respectively at the first category level. Specifically, the second category corresponding to the similar object at the specified category level may be determined according to the fourth category corresponding to any similar object at the first category level.

In practical applications, all trademarks and services are classified into 45 categories, forming "trademark registration supplies and service classifications", for example, the first category is chemicals, which may include "0101" industrial gas, "0102" industrial chemicals, the second category is color dyes, which may include "0201" dyes, the ninth category includes instruments, appliances, magnetic data carriers, which may include "0901" electronic computer and its peripherals, "0902" recording, vending machines, other counting detectors, and the like.

In the category scenario of trademark registration articles and services, the first category level may be a category level to which the categories such as the first category, the second category, or the ninth category in the foregoing examples belong, and the specified category may be a category in a category level below the first category, the second category, or the ninth category, for example, assuming that one category existing under the first category level is "2501" clothes, and categories at the second category level may also exist under the category, such as "250010" frock trousers, "250034" sweater, and other categories. This example is only for illustrating some examples of category division in the trademark scenario, and is not a specific limitation to the category division in the embodiment of the present application, and in practical applications, more division manners and division category levels may also be included, and specific categories corresponding to the respective category levels.

In a trademark application scene, under the condition that trademarks are similar to each other, the possibility that a newly applied trademark is authorized is low when the categories of the two trademarks are the same, and the possibility that the newly applied trademark is authorized is high when the categories of the two trademarks are different from each other. And if the similar object with the same first category as the object to be detected exists, the passing grade of the object to be detected can be kept unchanged. At this time, if present, the determining the target passage level based on the initial passage level may specifically include: and determining a target passing grade corresponding to the initial passing grade.

In this embodiment, a first category of an object to be detected and a second category corresponding to at least one similar object at an appointed category level are subjected to category matching, and a target category identical to the first category in the at least one second category is searched, so that the object to be detected and the at least one similar object are matched at the appointed category level, and the object to be detected is subjected to grouping category optimization matching through category inspection and judgment, so as to optimize the passing level of the object to be detected.

In order to further optimize the passing grade, the judgment of the user information can be carried out on the object to be detected and the similar object. In one possible design, determining the target pass level, if any, based on the initial pass level includes:

if yes, determining target user information of the target user and similar user information corresponding to at least one similar user respectively;

judging whether the at least one piece of similar user information comprises target user information;

if not, determining the initial passing grade as the target passing grade;

and if so, determining that the target passing level is the highest passing level in the at least one passing level corresponding to the grade division submodel.

In a trademark application scenario, the same user may apply for the same trademark multiple times, for example, multiple trademark applications in different trademark registration fields. If the user information is the same, the possibility that the newly applied trademark passes the examination is high. Therefore, when there is similar user information identical to the target user information, the highest passing level of the at least one passing level may be taken as the target passing level. In the embodiment, the incidence relation between the object to be detected and the identity information of the user is improved by detecting the identity information of the user, identity optimization of the user is provided, and accuracy of the target passing grade of the object to be detected is improved.

As another embodiment, the similar object detection submodel in the level prediction model is obtained by training:

determining at least one training object and reference results corresponding to the at least one training object respectively;

constructing a similar object detection sub-model;

and training the training targets with the same detection results of the similar object detection submodels on the respective training objects and the reference results respectively corresponding to the training to obtain model parameters of the similar object detection submodels.

In the embodiment of the application, a training mode of the similar object detection submodel is provided, and the model parameters of the similar object detection submodel can be obtained through training by using at least one training object and reference results respectively corresponding to the at least one training object, so that accurate model parameters are obtained, and the similar object detection submodel can be conveniently used.

The training of the similar object detection submodel to the training target with the same detection result of the at least one training and the reference result corresponding to the at least one training respectively may specifically include: initializing the model parameters of the similar object detection submodel to obtain reference model parameters; sequentially inputting at least one training object into the similar object detection submodel corresponding to the reference model parameter to obtain detection results corresponding to the at least one training object respectively; determining training errors based on the detection results respectively corresponding to the at least one training object and the reference results respectively corresponding to the at least one training object; judging whether the training error meets a convergence condition, if so, determining the reference model parameter as a model parameter; if not, updating the reference model parameter based on the training error, and returning to the step of sequentially inputting the similar object detection sub-models corresponding to the reference model parameter to the at least one training object to obtain the detection results respectively corresponding to the at least one training object to continue execution. The model parameters of the similar object detection submodel can be obtained through training in the training mode.

In a trademark detection scenario, determining at least one training object and reference results corresponding to the at least one training object respectively includes:

reading at least one trademark reject file;

analyzing at least one trademark rejection file to obtain at least one target trademark and at least one quotation trademark corresponding to the target trademark respectively;

and determining at least one target trademark as at least one training object, and respectively determining the quotation trademarks corresponding to the at least one target trademark as reference results corresponding to the at least one training object.

In a trademark application scenario, the training object and its corresponding reference result may be obtained by analyzing a trademark reject file. By analyzing the at least one reject file, the reference results corresponding to the at least one training object and the at least one training object can be obtained, and the analysis and the acquisition of the training objects are realized.

In some embodiments, the objects to be detected may include: and detecting character information or image information to be detected. At this time, the similar object detection model may include: a character detection unit and an image detection unit.

The inputting of the object to be detected into the similar object detection submodel to obtain at least one similar object matched with the object to be detected and the similarity between the at least one similar object and the object to be detected respectively may include:

inputting the character information to be detected into a character detection unit, detecting to obtain at least one character similar object and character similarity corresponding to the at least one character similar object respectively;

or, inputting the image to be detected into an image detection unit, detecting and obtaining at least one image similar object and the image similarity corresponding to the at least one image similar object respectively;

determining that at least one character similar object is at least one similar object, and the character similarity corresponding to the at least one character similar object is the similarity corresponding to the at least one similar object;

or determining that the at least one image similar object is the at least one similar object, and the image similarity corresponding to the at least one image similar object is the similarity corresponding to the at least one image similar object.

In practical application, the object to be detected may include characters and/or images, and particularly when the object to be detected is a trademark, the trademark may include characters and images, and the images are combined with different trademark styles in a certain combination manner based on the characters, so that detection of different contents of the object to be detected can be realized, and a more comprehensive detection scheme can be realized.

Additionally, in some embodiments, the objects to be detected may include: character information to be detected and image information to be detected. At this time, the similar object detection model may include: a character detection unit and an image detection unit. When the character information and the image are included at the same time, the similarity can be respectively identified and the similarity is fused to obtain accurate similarity, so that accurate target passing grade is obtained.

inputting the image to be detected into an image detection unit, detecting and obtaining at least one image similar object and the image similarity corresponding to the at least one image similar object respectively;

determining character object identifications corresponding to the at least one character similar object and image object identifications corresponding to the at least one image similar object;

determining a first object identifier with the same character object identifier and image object identifier;

determining a second object identifier except the first object identifier in the at least one character object identifier, and determining an object corresponding to a third object identifier of the first object identifier and the first object identifier in the at least one image object identifier as at least one similar object;

determining the comprehensive similarity corresponding to the first object identifier according to the character similarity and the image similarity corresponding to the first object identifier;

and determining the similarity corresponding to at least one object identifier according to the comprehensive similarity corresponding to the first object identifier, the character similarity corresponding to the second object identifier and the image similarity corresponding to the third object identifier.

Optionally, determining the comprehensive similarity corresponding to the first object identifier according to the similarity of the characters corresponding to the first object identifier and the similarity of the images may include: and selecting the character similarity corresponding to the first object identifier and the smaller similarity in the image similarity as the comprehensive similarity corresponding to the first object identifier.

Optionally, determining the comprehensive similarity corresponding to the first object identifier according to the similarity of the characters corresponding to the first object identifier and the similarity of the images may include: and carrying out weighted summation on the character similarity and the image similarity corresponding to the first object identifier to obtain the comprehensive similarity corresponding to the first object identifier.

As an embodiment, after the object to be detected is input into the similar object detection submodel, and the at least one similar object matched with the object to be detected and the similarity between the at least one similar object and the object to be detected are obtained, the method may further include:

and generating optimization prompt information based on the at least one similar object.

And outputting optimization prompt information for the target object.

Optimization hints may be generated based on the at least one similar object. After obtaining the at least one similar object, an optimization suggestion may be generated for the at least one similar object to prompt a user to optimize an object to be detected.

Optionally, generating the optimization hint information based on the at least one similar object may include: and respectively comparing the object to be detected with at least one similar object to obtain at least one comparison result, generating an adjustment suggestion by using the at least one comparison result, and generating optimization prompt information by using the adjustment suggestion.

Optionally, the generating of the optimization prompt information based on at least one similar object may further include: and generating an information prompt page in detail according to the detailed object information respectively corresponding to at least one similar object according to the sequence of the similarity from high to low, and taking the information prompt page as the optimized prompt information.

In yet another embodiment, generating the optimization hint information based on the at least one similar object may include: sending at least one similar object and an object to be detected to an optimization user, so that the optimization user generates an optimization suggestion for the object to be detected according to the at least one similar object and feeds back the optimization suggestion;

receiving the optimization suggestion for the optimization user feedback;

and generating the optimization prompt information corresponding to the optimization suggestion.

The optimization user can be an optimization equipment terminal of the third party optimization user, the optimization equipment terminal can receive at least one similar object and an object to be detected, display the at least one similar object and the object to be detected for the optimization user, detect an optimization suggestion of the optimization user on the object to be detected, and feed back the optimization suggestion obtained by detection to the computing equipment configured with the information processing method provided by the application.

Optionally, the optimization suggestion may include: at least one of a purchase suggestion, a modification suggestion, or a re-application suggestion.

As shown in fig. 3, a flowchart of another embodiment of an information processing method provided in an embodiment of the present application may include:

301: and acquiring the object to be detected provided by the target user and sent by the target client.

302: and inputting the object to be detected into a grade prediction model to obtain the initial passing grade of the object to be detected.

303: and performing grade optimization processing on the initial passing grade by using a grade optimization model to obtain the target passing grade of the object to be detected.

304: and generating grade prompt information based on the target passing grade.

305: and sending the grade prompt information to the target client side so that the target client side can display the grade prompt information and a target user can know the target passing grade.

In the embodiment of the application, the server can acquire the object to be detected provided by the target user and sent by the target client, so that the object to be detected is input into the grade prediction model, the initial passing grade of the object to be detected is acquired, and the preliminary prediction of the passing grade of the object to be detected is realized. And then, carrying out grade optimization processing on the initial passing grade by using a grade optimization model to obtain a target passing grade of the object to be detected, so as to realize further optimization processing on the passing grade and improve the precision of grade estimation. And sending the grade prompt information generated based on the target passing grade to the target user terminal, and displaying the grade prompt information by the target user terminal, so that the target user can know the target passing grade. In this embodiment, besides accurately obtaining the pass grade of the target user, the configuration of the client and the server is also adopted, and the object to be detected collected by the client is sent to the server for identification processing, so that the grade determination efficiency of the object to be detected is improved, the processing pressure of the client is reduced, the usability of the client is maintained, and the user experience is improved.

As an embodiment, generating the ranking hint information based on the target pass ranking may include:

at least one pass level in the level prediction model is determined.

Generating a grade bar control and a scale control corresponding to the grade bar control;

dividing the level bar control into at least one sub-control according to the sequence of the at least one passing level from low to high or from high to low based on the at least one passing level;

and adjusting the control position of the scale control to the position of the corresponding sub-control of the target passing level in the level bar control so as to obtain level prompt information.

As shown in fig. 4, a flowchart of another embodiment of an information processing method provided in an embodiment of the present application may include:

401: and acquiring the object to be detected provided by the target user.

402: and determining a plurality of third categories corresponding to the first category level.

The first category level may be the category level of the first level in the trademark domain, i.e. the highest category level. The first category level is higher than the specified category level in the foregoing embodiments.

A plurality of third categories may be included under the first category level. Any third category may be divided into a plurality of secondary categories according to the second category level under the first category level, that is, when the designated category level is the second category level, the designated category level may include a plurality of fourth categories.

403: and inputting the object to be detected into the grade prediction model corresponding to any third category to obtain the initial passing grade of the object to be detected corresponding to the third category so as to obtain the initial passing grade of the object to be detected corresponding to each of the third categories.

The different classes of level prediction models only have different searching ranges in the object prediction submodels, and the searching modes and the searching processes are the same. The hierarchical sub-models used for different categories are the same.

It should be noted that the object to be detected in the embodiment of the present application may have corresponding classification categories, and in the process of actually applying the level prediction model and the level optimization model to perform prediction optimization to obtain the target passing level of the object to be detected, the level prediction models in different categories may implement corresponding prediction, and the level optimization model is used to perform level optimization. That is, the specific manner of inputting the object to be detected into the class prediction model corresponding to any one of the third categories to obtain the initial passing class of the object to be detected corresponding to the third category is the same as that in the embodiment shown in fig. 2, the step of inputting the object to be detected into the similar object detection sub-model to obtain at least one similar object matched with the object to be detected and the similarity between the at least one similar object and the object to be detected, in addition, the class prediction models in different categories may be obtained by separate training, or may also be obtained by simultaneous training, and the prediction manner and the training process of the different class prediction models are the same as those in the embodiment shown in fig. 3, and are not described herein again.

404: and performing grade optimization processing on the initial passing grades respectively corresponding to the third categories by using a grade optimization model to obtain target passing grades respectively corresponding to the objects to be detected in the third categories.

405: and outputting the target passing grades respectively corresponding to the plurality of third categories for the target user.

In this embodiment, the object to be detected may correspond to each level of category, and when the object to be detected provided by the target user is obtained, a plurality of third categories corresponding to the first category may be determined. The plurality of third categories are category divisions at the first category level. When the passing grade of the object to be detected is determined, similarity matching can be performed on the object to be detected in each grade category. At this time, the object to be detected may be input into the level prediction model corresponding to any third category, and the initial passing level corresponding to the third category of the object to be detected is obtained, so as to obtain the initial passing levels of the object to be detected corresponding to the plurality of third categories, respectively. And respectively carrying out grade optimization processing on the initial passing grades corresponding to the third categories through the grade optimization model, so that target passing grades of the object to be detected corresponding to the third categories can be obtained, the initial passing grades of the third categories are obtained in a classified manner, multi-category passing grade analysis is realized, and the analysis efficiency is improved.

As an embodiment, outputting, for the target user, target passing levels respectively corresponding to the plurality of third categories may include:

sequencing the target passing grades respectively corresponding to the third categories according to the passing grades from high to low;

and sequentially outputting the target passing grades respectively corresponding to the plurality of third categories according to the sequence from high to low.

In the embodiment of the application, after the target passing levels respectively corresponding to the third categories are sorted from high to low according to the passing levels, the target passing levels corresponding to the third categories can be displayed according to the sorting order, so that the target passing levels are output according to the ranking order, and the prompt effect is improved.

As another embodiment, outputting, for the target user, target passage levels respectively corresponding to the plurality of third categories includes:

sequencing the target passing grades corresponding to the plurality of third categories according to the category sequence of the corresponding third categories;

and sequentially outputting the target passing grades corresponding to the third categories according to the order from high to low.

In the embodiment of the application, the target passing grades corresponding to the third categories are sorted according to the category sequence of the third categories, and then the target passing grades corresponding to the third categories are sequentially displayed according to the category sequence, so that sequential output according to the category sequence is realized, and the display effect is improved.

As another embodiment, after determining a plurality of third categories corresponding to the first category level, the method may further include:

a monitoring category selected by the target user from the plurality of third categories is detected.

Outputting the target passing grades respectively corresponding to the plurality of third categories for the target user may include:

determining a first display mode of the monitoring category and a second display mode of other third categories except the monitoring category in the plurality of third categories;

displaying a target passing grade corresponding to the monitoring category according to a preset first display mode;

and displaying the target passing grade of each of other third categories except the monitoring category in the plurality of third categories according to a preset second display mode.

The first display mode is different from the second display mode. The display effect of the first display mode is better than that of the second display mode, and the display effect of the first display mode is more prominent.

Further, optionally, the obtaining of the monitoring category selected by the target user from the plurality of third categories may include: and generating a plurality of category input prompt messages of the third category. Displaying category input prompt information for the target user. And acquiring the monitoring categories selected or input by the target user in the category input prompt information.

The user side of the target user can detect the third category selected or input by the target user in the prompting information input by the category, obtain the monitoring category and send the monitoring category and the object to be detected to the server. Category input prompt information may be used to prompt the target user to select or enter a third category that needs to be monitored.

The category input prompt message may be a command for providing a plurality of third categories of input and selection controls. The category input prompt may include a plurality of third categories.

Further, optionally, the first display mode may include: the display device comprises a first display shape, a first display position corresponding to the center of the first display shape and a first display color;

the second display mode may include: the second display shape, a second display position corresponding to the center of the second display shape and a second display color.

In this embodiment, the monitoring categories selected by the target user may be detected, and different display modes are adopted to display the passing levels obtained under the monitoring categories selected by the user and the passing levels under the categories not selected by the user, so as to distinguish the two display modes of the categories, display the monitoring categories selected by the target user in a focused manner, and achieve targeted display.

As another embodiment, after performing level optimization processing on the initial passing levels respectively corresponding to the plurality of third categories by using the level optimization model to obtain target passing levels respectively corresponding to the plurality of third categories for the object to be detected, the method may further include:

respectively generating respective category prompt information of the plurality of third categories based on the target passing grades of the object to be detected respectively corresponding to the plurality of third categories; and outputting category prompt information corresponding to the plurality of third categories respectively.

And any category prompt information is used for prompting the target passing grade of the object to be detected in the corresponding third category.

In practical application, category prompt information corresponding to the third categories can be output together for the target user to select to check the passing grades of the corresponding categories. Various output forms may be included in outputting the category prompt information.

The first output form: and outputting the category prompt information corresponding to the third categories in a control form, wherein a more common output mode can include that the third categories establish a topological structure in a node form, and each node is associated with a target passing level of the object to be detected corresponding to the third category, so that when a user clicks any node in the topological structure, the target passing level corresponding to the node is displayed. For easy understanding, referring to fig. 5, a manner of displaying category prompt information corresponding to a plurality of third categories is provided for the spherical topology structure according to the embodiment of the present application.

The second output form is: and respectively generating a grade prompting control for the plurality of third categories, associating a target passing grade corresponding to each grade prompting control with each grade prompting control, displaying the grade prompting controls respectively corresponding to the plurality of third categories according to a certain arrangement display mode, and displaying the corresponding target passing grade when a target user clicks any grade prompting control. The arrangement display mode for the field view may include: and arranging the level prompt controls in a circular ring form, and adjusting the size and the position of the controls according to the radius of the circular ring. For convenience of understanding, reference is made to fig. 6 for a display mode in which the category prompt information provided in the embodiment of the present application is a circular prompt control and circular prompt controls corresponding to a plurality of category prompt information are arranged in a ring form.

In order to enable the user to quickly find the target passing level corresponding to the required third category, in some of the above, outputting the category prompt information corresponding to each of the plurality of third categories may include:

displaying category prompt information corresponding to the plurality of third categories respectively to generate comprehensive prompt information; detecting the moving operation of the target user aiming at the comprehensive prompt information;

responding to the moving operation, and executing corresponding moving control processing on the plurality of third categories in the comprehensive prompt information to obtain the moved comprehensive prompt information;

and displaying the moved comprehensive prompt information.

Optionally, the comprehensive prompt information generated by the category prompt information corresponding to each of the plurality of third categories may be located in a central area of the display interface before movement. The moving operation may include: the method includes presetting movement types of a movement control, such as click operation, slide operation, drag operation, rotation operation or translation operation, and the movement control processing corresponding to the movement operation may specifically be identifying a movement track and a movement type of the movement operation and executing movement processing corresponding to the movement track and the movement type of the movement operation. And the mobile control is generated for the comprehensive prompt message and is used for executing the control operation which is predefined. Such as the more common zoom-in control or zoom-out control. When the target user clicks the mobile control, the integrated mobile information may be moved according to the mobile operation pre-associated with the mobile control.

Referring to fig. 5, the integrated hint information is displayed in the form of a spherical topology. The nodes respectively corresponding to the plurality of third categories can be category prompt information corresponding to the third categories, the topological structure is displayed in a spherical three-dimensional image form, a related three-dimensional moving control 501 can be arranged on the spherical topological structure for the convenience of moving and checking of a user, and when the user clicks or rotates the moving control, moving operation aiming at the comprehensive prompt information can be triggered.

Referring to fig. 6, the integrated prompt information is displayed in a circular ring structure, prompt controls of category prompt information corresponding to a plurality of third categories are displayed in a two-dimensional ring structure, and when the target user triggers the category prompt controls, detailed information of the third categories may be displayed in some embodiments. The detailed information of the third category may be displayed in the form of an information subpage when displayed.

In addition, the target user may also perform movement control of the general prompt information by movement operations such as dragging, and referring to fig. 5, assuming that the movement is a 45-degree sliding operation in the upper left corner in a display interface in which the general prompt information is displayed in the central area of the display interface 502 before the movement, movement processing corresponding to the 45-degree sliding operation may be performed, and specifically, referring to fig. 7, a display result of the general prompt information after the movement may be performed.

When the user views the category prompt information corresponding to each of the third categories, the category prompt information corresponding to each of the third categories may be moved. When the category prompt information corresponding to each of the plurality of third categories is moved, the category prompt information corresponding to each of the plurality of third categories may be moved as a whole. In this embodiment, category prompt information corresponding to each of the third categories is generated as a piece of comprehensive prompt information, so as to detect a movement operation of the target user for the comprehensive prompt information, and execute a movement control process corresponding to the movement operation, so that the target user can obtain the comprehensive prompt information displayed in different angles, directions and areas, thereby implementing display prompt of interactive category information and improving display effectiveness.

In some embodiments, when the target user triggers any category prompt message in the form of clicking, sliding through, or the like, the target user may be displayed with a target pass level corresponding to the third category in the category prompt message selected by the target user. The specific information can be displayed in the form of a prompt sub-page.

Outputting category prompt information corresponding to each of the third categories may include:

outputting category prompt pages corresponding to the category prompt information respectively corresponding to the third categories;

and detecting the selection operation triggered by the target user on any third category in the plurality of third categories to obtain the concerned category.

Generating a level prompt sub-page based on a target passing level corresponding to the target user output attention category;

and displaying the class prompt information of the level prompt sub-page for the target user.

In practical application, the display of the level prompt sub-page for the target user may specifically be performed by overlaying the display of the level prompt sub-page on the display pages of the category prompt information corresponding to the plurality of third categories, respectively. That is, at this time, outputting category prompt information corresponding to each of the plurality of third categories may specifically include: and outputting the category prompt pages corresponding to the category prompt information respectively corresponding to the third categories.

After the target user moves, the selection operation may be triggered for the category prompt information of any third category to obtain the category of interest. Referring to fig. 7, assuming that the target user triggers a selection operation for the category hint information of "39 transportation storage", a level hint sub-page 701 corresponding to the third category of the category hint information may be displayed.

It should be noted that, when the technical solution of the embodiment of the present application is configured in a server, for example, a cloud server, a prompt example or a page example such as fig. 5 to fig. 7 may be generated in the server and sent to a user side of a target user for display, and a specific generation process may refer to the description in the foregoing embodiment and is not described herein again.

As shown in fig. 8, a flowchart of another embodiment of an information processing method provided in an embodiment of the present application may include:

801: and acquiring the object to be detected provided by the target user.

802: and determining a plurality of third categories corresponding to the first category level.

803: and acquiring a target category selected by the target user from the plurality of third categories.

In practical application, the plurality of third categories may be respectively prompted so that the user may select a target category that needs level confirmation. The plurality of third categories may be used as prompt nodes to prompt in an output form of a topological structure, and may also be used as prompt objects to generate a prompt control, such as a circular control, a spherical control, or a polygonal control, respectively, and each prompt control is used to prompt the corresponding third category. In addition, to improve the effectiveness of the display, the target category may be highlighted. For example, the shape and color of the prompt control of the target class selected by the user are set to be more prominent than the shape and color of the rest prompt controls, so that the prompt effect is improved.

804: and inputting the object to be detected into a grade prediction model corresponding to the target category to obtain the initial passing grade of the object to be detected in the target category.

805: and performing grade optimization processing on the initial passing grade by using a grade optimization model to obtain the target passing grade of the object to be detected.

806: and outputting the target passing grade corresponding to the target category for the target user.

In this embodiment, the object to be detected provided by the target user is obtained, the plurality of third categories respectively corresponding to the first category level may be determined, and the target category selected by the target user from the plurality of third categories may be obtained. The target category can be a category which is focused by a target user, through the target category setting, the targeted passing grade estimation can be carried out, an accurate grade estimation target is obtained, and the effectiveness of the grade estimation is improved.

The obtaining a target category selected by the target user from the plurality of third categories comprises:

generating category input prompt information of the plurality of third categories;

displaying the category input prompt information for the target user;

and acquiring the target category selected or input by the target user in the category input prompt information.

Alternatively, the category input prompt message may be a category input text input control. The user may enter the target category directly in the category entry text entry control. The text input control may further include a selection prompt sub-control corresponding to each of the plurality of third categories, and when the text input control is selected by the detection target user cursor, the selection prompt sub-controls corresponding to each of the plurality of third categories may be displayed for the target user to prompt the target user to select the category, and the target category selected by the target user is detected and obtained.

Taking an object to be detected as an example of a trademark to be detected, fig. 9 shows a flowchart of another embodiment of an information processing method provided in an embodiment of the present application, where the method may include:

901: and acquiring the trademark to be detected provided by the target user.

902: inputting the trademark to be detected into the grade prediction model to obtain the initial passing grade of the trademark to be detected.

903: and performing grade optimization processing on the initial passing grade by using a grade optimization model to obtain the target passing grade of the trademark to be detected.

904: and outputting the target passing grade for the target user.

In this embodiment, the trademark to be detected of the target user may be acquired, and then, the trademark to be detected may be input to the grade prediction model to acquire the initial passing grade of the trademark to be detected. At this time, the passing grade of the trademark to be detected is preliminarily predicted by using the grade prediction model. And then, carrying out grade optimization processing on the initial passing grade by using a grade optimization model to obtain the target passing grade of the trademark to be detected. At the moment, the grade optimization model is utilized to further grade optimize the initially predicted passing grade, and the target passing grade accurately measured by the trademark to be detected is obtained, so that when the target user outputs the target passing grade, the passing possibility of the trademark to be detected of the target user can be effectively prompted through the accurate target passing grade, and the generation of invalid trademarks is avoided. Particularly, when the trademark to be detected is the trademark, the success rate of trademark registration can be improved.

For convenience of understanding, the technical solutions of the embodiments of the present application are described in detail by taking the object to be detected as a trademark as an example.

In fig. 10, the target user can input the trademark to be detected using the user terminal M1. In one possible design, the user end M1 may display an input interface of a trademark to be detected, and the target user U1 may input the textual information and the image information of the trademark in the textual input control 1001 and the image input control 1002 provided in the input interface. At this time, the user M1 may detect the trademark to be detected, which is formed by the text information and the image information input by the target user, and send the trademark to be detected to the server M2.

In addition, in some embodiments, the target user may further specifically define the information type of the trademark, and may specifically provide prompt information of a candidate information type, such as a prompt control shown in 1003, which is combined by text, graphics, or graphics, and the user may select the information type of the trademark by triggering the prompt control. In addition, in some other embodiments, the target user may select a monitoring category from a plurality of third categories, where an input prompt control 1004 for the selected category may be displayed in the trademark input interface, and the user may trigger the selection control 1004 to select the corresponding monitoring category.

The server M2 may acquire the trademark to be detected provided by the target user, which may include, for example, text information of the trademark and image information of the trademark. The server M2 may then input the trademark to be detected into the grade prediction model, and obtain the initial passing grade of the object to be detected. After the initial passing grade is obtained, the grade optimization model can be used for carrying out grade optimization processing on the initial passing grade to obtain the target passing grade of the trademark to be detected. Thereafter, a target pass rating may be output for the target user.

In practical applications, the server M2 outputs the target passing level to the user terminal M1, and the display of the user terminal M1 outputs the target passing level. When the target passing level is output, the target passing level may be output in the form of a level bar control, and specifically, the level bar control 1005 shown in fig. 7 may be referred to.

In addition, in some embodiments, the passing grades of the trademark to be detected under a plurality of third categories may be respectively predicted, the passing grades of the trademark to be detected under different third categories are obtained, and the passing grades are respectively prompted. The prompt information 1006 of the target passing grades respectively corresponding to the trademarks to be detected under a plurality of third categories is shown in fig. 10. The plurality of objects shown in 1006 are presented in order of category by the hint information of the hierarchy. Further, 1006 is a hint information of a plurality of target passing levels under a fourth category at a second category level under the first category level where the third category is located.

In addition, in some embodiments, category prompt information corresponding to trademarks in a plurality of third categories may be displayed in a spherical topology, referring to category prompt information 1007 arranged in a circular ring form in fig. 10. In the prompt information 1007, the category prompt information of the third category "20 furniture items" selected by the target user is highlighted to improve the prompt effect.

As shown in fig. 11, a schematic structural diagram of an embodiment of a computing device provided in an embodiment of the present application, where the computing device may include: a storage component 1101 and a processing component 1102; storage component 1101 is used to store one or more computer instructions; one or more computer instructions are invoked and executed by the processing component 1102;

the processing component 1102 is configured to:

As an embodiment, the inputting, by the processing component, the object to be detected into the level prediction model, and the obtaining of the initial passing level of the object to be detected may specifically include:

inputting the object to be detected into a similar object detection submodel to obtain at least one similar object matched with the object to be detected and the similarity between the at least one similar object and the object to be detected respectively;

and based on the similarity corresponding to at least one similar object, carrying out grading processing according to the grading sub-model to obtain the initial passing grade of the object to be detected.

In some embodiments, the performing, by the processing component, a ranking process according to the ranking submodel based on the similarity corresponding to each of the at least one similar object, and the obtaining an initial passing rank of the object to be detected may specifically include:

acquiring at least one passing grade corresponding to the grade division submodel and a similarity range corresponding to each of the at least one passing grade;

In one possible design, the determining, by the processing component, a target similarity range in the at least one similarity range based on the respective similarities of the at least one similar object may specifically include:

and determining a target similarity range corresponding to the maximum similarity in the at least one similarity range.

In another possible design, the determining, by the processing component, the target similarity range in the at least one similarity range based on the similarity corresponding to the at least one similar object may specifically include:

and determining a target similarity range with the highest range in the at least one candidate similarity range.

In some embodiments, the performing, by the processing component, a level optimization process on the initial pass level by using a level optimization model, and the obtaining of the target pass level of the object to be detected may specifically include:

if so, determining a target pass level based on the initial pass level;

and if the target passing level does not exist, determining that the target passing level is the highest passing level in at least one passing level corresponding to the level division submodel.

In one possible design, the processing component, if present, determining the target pass level based on the initial pass level may specifically include:

if the object exists, determining the target user information of the target user and the similar user information corresponding to the at least one similar object respectively;

if not, determining the initial passing grade as the target passing grade;

and if so, determining the target passing level as the highest passing level in at least one passing level corresponding to the grade division submodel.

As yet another embodiment, the processing component trains the similar object detection submodel in the level prediction model by:

constructing a similar object detection sub-model;

and training the training targets with the same detection results of the similar object detection submodels on the respective training targets as the reference results corresponding to the at least one training target respectively to obtain model parameters of the similar object detection submodels.

Further, optionally, the determining, by the processing component, the at least one training object and the reference result corresponding to the at least one training object respectively may specifically include:

reading at least one trademark reject file;

In one possible design, the object to be detected includes: character information or image information to be detected; the similar object detection submodel includes: a character detection unit and an image detection unit;

the inputting, by the processing component, the object to be detected into the similar object detection submodel, and the obtaining of the at least one similar object matched with the object to be detected and the similarity between the at least one similar object and the object to be detected may specifically include:

or inputting the information of the image to be detected into an image detection unit, and detecting to obtain at least one image similar object and the image similarity corresponding to the at least one image similar object respectively;

determining at least one similar object corresponding to at least one character similar object, wherein the character similarity corresponding to at least one character similar object is the similarity corresponding to at least one similar object;

or determining at least one similar object corresponding to the at least one image similar object, and determining the image similarity corresponding to the at least one image similar object respectively as the similarity corresponding to the at least one similar object respectively.

In yet another possible design, the object to be detected includes: character information and image information to be detected are detected;

inputting the information of the image to be detected into an image detection unit, detecting and obtaining at least one image similar object and the image similarity corresponding to the at least one image similar object respectively;

determining the object identification of at least one character similar object and the object identification of at least one image similar object;

carrying out object duplicate removal processing based on the object identifier of at least one character similar object and the object identifier of at least one image similar object to obtain at least one similar object;

determining the similarity of similar objects with the same object identification according to the character similarity of the character similar objects with the same object identification and the image similarity of the image similar objects;

and taking the character similarity of the character similar objects of which the object identifications are not overlapped as the similarity of the corresponding similar objects and the image similarity of the image similar objects of which the object identifications are not overlapped as the similarity of the corresponding similar objects to obtain the object similarity of at least one similar object.

In some embodiments, the processing component may be further to:

generating optimization prompt information based on at least one similar object;

and outputting optimization prompt information for the target object.

As a possible implementation manner, the processing component generating the optimization hint information based on the at least one similar object may include:

the at least one similar object and the object to be detected are sent to an optimization user, so that the optimization user can generate an optimization suggestion for the object to be detected according to the at least one similar object and feed back the optimization suggestion;

receiving the optimization suggestion for the optimization user feedback;

and generating optimization prompt information corresponding to the optimization suggestion.

As another embodiment, the acquiring, by the processing component, the object to be detected provided by the target user may specifically include:

acquiring an object to be detected provided by a target user and sent by a target client;

outputting the target pass rating for the target user includes:

generating grade prompt information based on the target passing grade;

and sending the grade prompt information to the target client side so that the target client side can display the grade prompt information and a target user can know the target passing grade.

In some embodiments, the generating, by the processing component, the level hint information based on the target pass level may specifically include:

determining at least one pass level in a level prediction model;

dividing the level bar control into at least one sub-control according to the sequence of the at least one passing level from low to high or from high to low on the basis of the at least one passing level;

In some embodiments, the inputting, by the processing component, the object to be detected into the level prediction model, and the obtaining the initial passing level of the object to be detected may specifically include:

determining a plurality of third categories corresponding to the first category level;

inputting the object to be detected into a grade prediction model corresponding to any third category to obtain an initial passing grade of the object to be detected corresponding to the third category so as to obtain initial passing grades of the object to be detected corresponding to a plurality of third categories respectively;

the processing component performs level optimization processing on the initial pass level by using a level optimization model, and obtaining a target pass level of the object to be detected may specifically include:

respectively carrying out grade optimization processing on the initial passing grades respectively corresponding to the third categories by utilizing a grade optimization model to obtain target passing grades respectively corresponding to the objects to be detected in the third categories;

the outputting, by the processing component, the target pass level for the target user may specifically include:

and outputting the target passing grades respectively corresponding to the plurality of third categories for the target user.

In a possible design, the outputting, by the processing component, target passing levels respectively corresponding to the plurality of third categories for the target user may specifically include:

In another possible design, the outputting, by the processing component, target passing levels respectively corresponding to the plurality of third categories for the target user may specifically include:

As a possible implementation, the processing component may be further configured to:

detecting a monitoring category selected by the target user from the plurality of third categories;

the step of outputting, by the processing component, target passing levels respectively corresponding to the plurality of third categories for the target user may specifically include:

determining a first display mode of the monitoring category and a second display mode of each of other categories except the monitoring category in the plurality of third categories;

and displaying the target passing grade of each of the other categories except the monitoring category in the plurality of third categories according to a preset second display mode.

In some embodiments, the obtaining, by the processing component, the monitoring category selected by the target user from the plurality of third categories may specifically include:

generating category input prompt information of a plurality of third categories;

displaying category input prompt information for a target user;

and acquiring the monitoring categories selected or input by the target user in the category input prompt information.

Further, optionally, the first display mode includes: the display device comprises a first display shape, a first display position corresponding to the center of the first display shape and a first display color;

the second display mode includes: the second display shape, a second display position corresponding to the center of the second display shape and a second display color.

In still other embodiments, the processing component may be further to:

respectively generating category prompt information of the plurality of third categories based on target passing grades of the object to be detected respectively corresponding to the plurality of third categories;

outputting category prompt information corresponding to the plurality of third categories respectively;

the outputting, by the processing component, category prompt information corresponding to each of the plurality of third categories may specifically include:

and displaying the moved comprehensive prompt information.

In some embodiments, the outputting, by the processing component, category prompt information corresponding to each of the plurality of third categories may specifically include:

detecting a selection operation triggered by the target user aiming at any third category in the plurality of third categories to obtain a concerned category;

outputting a target passing grade corresponding to the attention category based on the target user, and generating a grade prompt sub-page;

and displaying the grade prompt sub-page for the target user.

As yet another example, the processing group may be further to:

acquiring a target category selected by the target user from a plurality of third categories;

the inputting, by the processing component, the object to be detected into the level prediction model, and the obtaining of the initial passing level of the object to be detected may specifically include:

inputting the object to be detected into a grade prediction model corresponding to the target category to obtain the initial passing grade of the object to be detected in the target category;

and outputting the target passing grade corresponding to the target category for the target user.

As an embodiment, the obtaining, by the processing component, a target category selected by the target user from the plurality of third categories may specifically include:

displaying the category input prompt information for the target user;

The implementation computing device of fig. 11 may execute the method for processing information of the embodiment shown in fig. 1, and details of implementation principles and technical effects are not repeated. The specific manner in which the various steps are performed by the processing elements in the above-described embodiments has been described in detail in relation to embodiments of the method and will not be set forth in detail herein.

In addition, an embodiment of the present application further provides a computer-readable storage medium, where the storage medium is used to store a computer program, and when the computer program is executed, the information processing method in the embodiment shown in fig. 1 may be executed.

As shown in fig. 12, a schematic structural diagram of another embodiment of a computing device provided in an embodiment of the present application may include: a storage component 1201 and a processing component 1202; storage component 1201 is used to store one or more computer instructions; one or more computer instructions are invoked and executed by the processing component 1202;

the processing component 1202 may be configured to:

The implementation computing device in fig. 12 may execute the method for processing information in the embodiment shown in fig. 9, and details of implementation principles and technical effects are not repeated. The specific manner in which the various steps are performed by the processing elements in the above-described embodiments has been described in detail in relation to embodiments of the method and will not be set forth in detail herein.

In addition, an embodiment of the present application further provides a computer-readable storage medium, where the storage medium is used to store a computer program, and when the computer program is executed, the information processing method in the embodiment shown in fig. 6 may be executed.

The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by adding a necessary general hardware platform, and of course, can also be implemented by a combination of hardware and software. With this understanding in mind, the above-described technical solutions and/or portions thereof that contribute to the prior art may be embodied in the form of a computer program product, which may be embodied on one or more computer-usable storage media having computer-usable program code embodied therein (including but not limited to disk storage, CD-ROM, optical storage, etc.).

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.

Claims

1. An information processing method characterized by comprising:

acquiring an object to be detected provided by a target user;

and outputting the target passing grade for the target user.

2. The method according to claim 1, wherein the inputting the object to be detected into a grade prediction model, and obtaining the initial passing grade of the object to be detected comprises:

and based on the similarity corresponding to the at least one similar object, carrying out grading processing according to a grading sub-model to obtain the initial passing grade of the object to be detected.

3. The method according to claim 2, wherein the obtaining of the initial pass level of the object to be detected by performing level division processing according to a level division submodel based on the respective corresponding similarity of the at least one similar object comprises:

acquiring at least one passing grade corresponding to the grade division submodel and a similarity range corresponding to the at least one passing grade respectively;

determining a target similarity range in at least one similarity range based on the similarity corresponding to the at least one similar object respectively;

and determining the passing grade corresponding to the target similarity range as the initial passing grade.

4. The method according to claim 3, wherein the determining a target similarity range in at least one similarity range based on the respective similarities of the at least one similar object comprises:

5. The method according to claim 3, wherein the determining a target similarity range in at least one similarity range based on the respective similarities of the at least one similar object comprises:

and determining the target similarity range with the highest range in the at least one candidate similarity range.

6. The method according to claim 2, wherein the performing level optimization processing on the initial pass level by using a level optimization model to obtain a target pass level of the object to be detected comprises:

determining a first category corresponding to the object to be detected at a specified category level;

determining second categories corresponding to the at least one similar object respectively at the specified category level;

judging whether a target category which is the same as the first category exists in second categories respectively corresponding to the at least one similar object;

if so, determining the target pass level based on the initial pass level;

7. The method of claim 6, wherein determining the target pass level, if any, based on the initial pass level comprises:

if so, determining the target user information of the target user and the similar user information corresponding to the at least one similar object respectively;

judging whether the at least one piece of similar user information comprises the target user information;

if not, determining that the initial passing grade is the target passing grade;

if so, determining the target passing level as the highest passing level in at least one passing level corresponding to the level division submodel.

8. The method of claim 2, wherein the similar object detection submodel in the level prediction model is obtained by training:

constructing a similar object detection sub-model;

and training the training targets with the same detection results of the similar object detection submodels as the reference results corresponding to the at least one training object to obtain model parameters of the similar object detection submodels.

9. The method of claim 8, wherein determining the at least one training object and the reference result corresponding to the at least one training object respectively comprises:

reading at least one trademark reject file;

analyzing the at least one trademark rejection file to obtain at least one target trademark and quotation trademarks corresponding to the at least one target trademark respectively;

and determining that the at least one target trademark is the at least one training object, and the quotation trademarks respectively corresponding to the at least one target trademark are reference results respectively corresponding to the at least one training object.

10. The method according to claim 2, wherein the object to be detected comprises: character information or image information to be detected; the similar object detection submodel includes: a character detection unit and an image detection unit;

the inputting the object to be detected into a similar object detection submodel to obtain at least one similar object matched with the object to be detected and the similarity between the at least one similar object and the object to be detected respectively comprises:

inputting the character information to be detected into the character detection unit, detecting to obtain at least one character similar object and character similarity corresponding to the at least one character similar object respectively;

or inputting the image information to be detected into the image detection unit, and detecting to obtain at least one image similar object and the image similarity corresponding to the at least one image similar object;

determining at least one similar object corresponding to the at least one similar literal object, and determining the respective corresponding literal similarities of the at least one similar literal object as the respective corresponding similarities of the at least one similar literal object;

or determining at least one similar object corresponding to the at least one image similar object, and determining the image similarity corresponding to the at least one image similar object as the similarity corresponding to the at least one similar object.

11. The method of claim 10, wherein the object to be detected comprises: character information and image information to be detected are detected;

inputting the information of the image to be detected into the image detection unit, detecting and obtaining at least one image similar object and the image similarity corresponding to the at least one image similar object respectively;

determining the object identification of the at least one character similar object and the object identification of the at least one image similar object;

performing object de-duplication processing on the basis of the object identifier of the at least one character similar object and the object identifier of the at least one image similar object to obtain at least one similar object;

determining the similarity of similar objects with the same object identification according to the character similarity of similar objects with the same object identification and the image similarity of similar objects with the same image identification;

and taking the character similarity of the character similar objects with non-overlapping object identifications as the similarity of the corresponding similar objects and the image similarity of the image similar objects with non-overlapping object identifications as the similarity of the corresponding similar objects to obtain the object similarity of the at least one similar object.

12. The method of claim 2, further comprising:

generating optimization prompt information based on the at least one similar object;

and outputting the optimization prompt information for the target object.

13. The method of claim 12, wherein generating optimization hint information based on the at least one similar object comprises:

receiving the optimization suggestion for the optimization user feedback;

14. The method according to claim 1, wherein the acquiring the object to be detected provided by the target user comprises:

acquiring an object to be detected provided by the target user and sent by a target client;

the outputting the target pass rating for the target user comprises:

generating grade prompt information based on the target passing grade;

and sending the grade prompt information to the target client side so that the target client side can display the grade prompt information and the target user can know the target passing grade.

15. The method of claim 14, wherein generating a rating hint based on the target pass rating comprises:

determining at least one pass level in the level prediction model;

and adjusting the control position of the scale control to the corresponding sub-control of the target passing grade in the grade bar control so as to obtain the grade prompt information.

16. The method according to claim 1, wherein the inputting the object to be detected into a grade prediction model, and obtaining the initial passing grade of the object to be detected comprises:

performing level optimization processing on the initial passing level by using a level optimization model to obtain a target passing level of the object to be detected comprises:

the outputting the target pass rating for the target user comprises:

17. The method according to claim 16, wherein the outputting the target passing grades respectively corresponding to the plurality of third categories for the target user comprises:

18. The method according to claim 16, wherein the outputting the target passing grades respectively corresponding to the plurality of third categories for the target user comprises:

19. The method of claim 16, further comprising:

the outputting the target passing grades respectively corresponding to the plurality of third categories for the target user comprises:

determining a first display mode of the monitoring category and a second display mode of each of other third categories except the monitoring category in the plurality of third categories;

displaying the target passing grade corresponding to the monitoring category according to a preset first display mode;

20. The method of claim 19, wherein the first display mode comprises:

the display device comprises a first display shape, a first display position corresponding to the center of the first display shape and a first display color;

the second display mode includes: a second display shape, a second display position corresponding to the center of the second display shape, and a second display color.

21. The method of claim 16, further comprising:

and outputting category prompt information corresponding to the plurality of third categories respectively.

22. The method according to claim 21, wherein the outputting category prompt information corresponding to each of the plurality of third categories comprises:

and displaying the moved comprehensive prompt information.

23. The method according to claim 21, wherein the outputting category prompt information corresponding to each of the plurality of third categories comprises:

and displaying the grade prompt sub-page for the target user.

24. The method of claim 1, further comprising:

the inputting the object to be detected into a grade prediction model to obtain an initial passing grade of the object to be detected comprises:

the outputting the target pass rating for the target user comprises:

25. The method of claim 24, wherein the obtaining a target category selected by the target user from the plurality of third categories comprises:

displaying the category input prompt information for the target user;

26. An information processing method characterized by comprising:

acquiring a trademark to be detected provided by a target user;

and outputting the target passing grade for the target user.

27. A computing device, comprising: a storage component and a processing component; the storage component is used for storing one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component;

the processing component is to:

28. A computing device comprising a storage component and a processing component; the storage component is used for storing one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component;

the processing component is to: