Information Push Method, Device, Electronic Device and Storage Medium

Abstract

Embodiments of the present application disclose an information push method, apparatus, electronic device, and storage medium. The method includes: obtaining the assessment situations of multiple assessors for the assessment of the titles of pet veterinarians; determining the feature vectors of each of the multiple assessors based on the assessment situations of each of the multiple assessors; clustering the multiple assessors according to the feature vectors of each of the assessors to obtain at least one clustering result; and pushing push information corresponding to each clustering result to the assessors included in each of the at least one clustering results. Embodiments of the present application are beneficial to improving the utilization rate of assessment results.

Description

Technical Field

[0001] This application relates to the technical field of big data processing, and in particular, to an information push method, apparatus, electronic device, and storage medium. Background Art

[0002] Currently, in order to simplify the assessment process, various online assessment systems have been developed. For example, the existing pet doctor title assessment system can linearly assess the assessors to determine whether they pass the pet doctor title assessment. However, various current online assessment systems only simply display the assessment results of the assessors, resulting in relatively low utilization rate of the assessment results and waste of resources. Summary of the Invention

[0003] Embodiments of this application provide an information push method, apparatus, electronic device, and storage medium. Information is pushed to the assessors based on the assessment results of the pet doctor title, improving the utilization rate of the assessment results and avoiding waste of resources.

[0004] In a first aspect, an embodiment of this application provides an information push method, including:

[0005] Obtain the assessment situations of multiple assessors for the pet doctor title;

[0006] Based on the assessment situation of each assessor among the multiple assessors, determine the feature vector of each assessor;

[0007] Cluster the multiple assessors according to the feature vector of each assessor to obtain at least one clustering result;

[0008] Push the push information corresponding to each clustering result to the assessors included in each clustering result among the at least one clustering result.

[0009] In a second aspect, an embodiment of this application provides an information push apparatus, including: an obtaining unit and a processing unit;

[0010] The obtaining unit is configured to obtain the assessment situations of multiple assessors for the pet doctor title;

[0011] The processing unit is configured to determine the feature vector of each assessor based on the assessment situation of each assessor among the multiple assessors;

[0012] Cluster the multiple assessors according to the feature vector of each assessor to obtain at least one clustering result;

[0013] Push the push information corresponding to each clustering result to the assessors included in each clustering result among the at least one clustering result.

[0014] In a third aspect, an embodiment of the present application provides an electronic device, including: a processor, the processor is connected to a memory, the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory so that the electronic device executes the method described in the first aspect.

[0015] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and the computer program enables a computer to execute the method described in the first aspect.

[0016] In a fifth aspect, an embodiment of the present application provides a computer program product, where the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer is operable to enable a computer to execute the method described in the first aspect.

[0017] Implementing the embodiments of the present application has the following beneficial effects:

[0018] It can be seen that in the embodiments of the present application, after an examiner participates in the assessment of a pet doctor title through an online assessment system, the assessment situations of multiple examiners can be obtained, and the multiple examiners can be clustered based on the assessment situations of each examiner, and then push the push information matching the clustering result to the examiners in each clustering result, so as to realize intelligent and personalized information push through the assessment situation of the pet doctor title, improving the assessment experience of the examiners; and it is not simply to display the assessment situations of the examiners, improving the utilization rate of the resource of the assessment situations of the examiners. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0020] Figure 1 It is a schematic diagram of an information push system provided by an embodiment of the present application;

[0021] Figure 2 It is a schematic flowchart of an information push method provided by an embodiment of the present application;

[0022] Figure 3 It is a schematic diagram of obtaining a feature vector of each examiner provided by an embodiment of the present application;

[0023] Figure 4 It is a schematic flowchart of obtaining the assessment situation of an examiner provided by an embodiment of the present application;

[0024] Figure 5 A schematic diagram for intercepting at least one first voice segment provided by an embodiment of the present application;

[0025] Figure 6 A schematic diagram for traversing assessment items and first text content provided by an embodiment of the present application;

[0026] Figure 7 A functional unit composition block diagram of an information push device provided by an embodiment of the present application;

[0027] Figure 8 A schematic structural diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0028] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are some, rather than all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art without making creative efforts based on the embodiments in the present application belong to the scope of protection of the present application.

[0029] The terms "first", "second", "third", "fourth", etc. in the specification and claims of the present application and the accompanying drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products, or devices.

[0030] Referring to

[0031] Referring to Figure 1 , Figure 1 A schematic diagram of an information push system provided by an embodiment of the present application. The information push system includes multiple user terminals 10 and an information push device 20.

[0032] Exemplarily, the information push device 20 obtains the assessment situations of multiple assessors for the assessment of the pet doctor title; based on the assessment situations of each assessor among the multiple assessors, determines the feature vector of each assessor; according to the feature vector of each assessor, clusters the multiple assessors to obtain at least one clustering result. Then, pushes the push information corresponding to the clustering result to the user terminals 10 of the assessors included in each clustering result among the at least one clustering result. Exemplarily, the push information may be training information for the pet doctor title, or social information, etc.

[0033] It can be seen that in the embodiment of the present application, after the assessor completes the assessment of the pet doctor title through the online assessment system, the assessment situations of multiple assessors can be obtained, and the multiple assessors are clustered based on the assessment situations of each assessor, and then the push information matching the clustering result is pushed to the assessors in each clustering result, so as to realize intelligent and personalized information push through the assessment situation of the pet doctor title, improve the assessment experience of the assessors; and it is not simply to display the assessment situations of the assessors, improving the utilization rate of the resource of the assessment situations of the assessors.

[0034] Refer to Figure 2 , Figure 2 is a schematic flowchart of an information push method provided by an embodiment of the present application. This method is applied to the above-mentioned information push device 20. This method includes the following steps:

[0035] 201: Obtain the assessment situations of multiple assessors for the assessment of the pet doctor title.

[0036] Exemplarily, each assessor can assess the pet doctor title through the online assessment system. For example, the assessor can assess the pet attending doctor through the online assessment system, or assess the assistant of the pet attending doctor through the online assessment system, etc. It should be noted that N assessment items, that is, N test questions, are set for the assessment of each pet doctor title, and N is an integer greater than or equal to 1.

[0037] Optionally, the assessment situation of each assessor in the present application includes but is not limited to:

[0038] The answering duration of the assessor for each of the N assessment items (that is, the answering duration for each test question), the number of modifications for each assessment item, and the score obtained for each assessment item (that is, the assessment result of each assessment item). The specific implementation process of obtaining the assessment situation of each assessor is described below and will not be described in detail here.

[0039] 202: Based on the assessment situations of each assessor among the multiple assessors, determine the feature vector of each assessor.

[0040] Exemplarily, based on the response time of each appraiser for each of the N appraisal items, a first feature vector corresponding to each appraiser is constructed; based on the number of modifications of each appraiser for each of the N appraisal items, a second feature vector corresponding to each appraiser is constructed; based on the score obtained by each appraiser for each of the N appraisal items, a third feature vector corresponding to each appraiser is constructed.

[0041] Exemplarily, as Figure 3 shown, according to the appraisal order (i.e., question number) of the N appraisal items, the response times of each appraiser for each of the N appraisal items are sorted to obtain a first feature vector corresponding to each appraiser. Similarly, according to the appraisal order of the N appraisal items, the number of modifications of each appraiser for each appraisal item is sorted to obtain a second feature vector, and the scores obtained by each appraiser for each appraisal item are sorted to obtain a third feature vector.

[0042] Further, the first feature vector, the second feature vector, and the third feature vector of each appraiser are concatenated to obtain the feature vector of each appraiser. As Figure 3 shown, the first feature vector, the second feature vector, and the third feature vector of each appraiser are vertically concatenated to obtain the feature vector of each appraiser.

[0043] 203: According to the feature vectors of each appraiser, the multiple appraisers are clustered to obtain at least one clustering result.

[0044] Exemplarily, the feature vectors of each appraiser can be clustered through a clustering algorithm to obtain at least one clustering result, where each clustering result corresponds to a clustering center, and each clustering center is characterized by a feature vector composed of response time, number of modifications, and obtained score. That is to say, each clustering center also represents an appraisal situation.

[0045] Optionally, the similarity between any two feature vectors can be calculated, and based on this similarity, it is determined whether the two feature vectors can be grouped into one category, so as to realize the clustering of multiple appraisals.

[0046] 204: Push the push information corresponding to each clustering result to the appraisers included in each clustering result among the at least one clustering result.

[0047] Exemplarily, according to the clustering center of each clustering result, the push information corresponding to each clustering result is determined, and the push information is pushed to the appraisers included in each clustering result.

[0048] Specifically, according to the cluster centers of each clustering result, they are respectively matched with the feature vectors of each candidate information in the information library to obtain the matching degree corresponding to each candidate information. The candidate information with a matching degree greater than the threshold is used as the push information corresponding to the clustering result. Among them, the feature vector of each candidate information is obtained by pre-vectorizing the features of each candidate information. Exemplarily, the candidate information includes training course information corresponding to each pet doctor title, teaching video information corresponding to each pet doctor title, and social information corresponding to each pet doctor title, such as social locations, social official accounts, and so on.

[0049] It can be seen that in the embodiment of the present application, after the examiner participates in the assessment of the pet doctor title through the online assessment system, the assessment situations of multiple examiners can be obtained, and the multiple examiners are clustered based on the assessment situations of each examiner. Then, push information matching the clustering result is pushed to the examiners in each clustering result, so as to realize intelligent and personalized information push through the assessment situations of the pet doctor title, improving the assessment experience of the examiners; moreover, it is not simply to display the assessment situations of the examiners, improving the utilization rate of the resource of the assessment situations of the examiners.

[0050] In the present application, an example is mainly given where the examiner conducts the assessment of the pet doctor title in the way of voice assessment. Other assessment modes are similar and will not be described.

[0051] Refer to Figure 4 , Figure 4 FIG. is a schematic flowchart of a method for obtaining the assessment situation of an examiner provided by an embodiment of the present application. This method is applied to the above information push device. This method includes the following steps:

[0052] 401: Obtain the assessment voice of each examiner among the multiple examiners.

[0053] Among them, the assessment voice of each examiner is the assessment voice for the assessment of this pet doctor title.

[0054] Among them, the assessment voice of each examiner is obtained from the voice assessment system, that is, the multiple examiners are uniformly and online assessed through the voice assessment system, and the assessment voice of each examiner is saved.

[0055] 402: Determine the answering duration of each examiner for each assessment item among the N assessment items according to the assessment voice of each examiner.

[0056] First, it should be noted that the answering duration mentioned in the present application refers to the duration used to answer each assessment item for the first time, without calculating the duration used for subsequent modification or adjustment of this assessment item.

[0057] It is understandable that the total assessment time of the assessor is fixed, and the assessment time for each assessment item is also preset in advance. Then, the voice assessment system will call the question voice of each assessment item to the assessor within the preset assessment time; then, a corresponding response time period will be reserved for each assessment item, and the assessor can answer the assessment item within this response time period. However, within this response time period, the assessor may only use a part of the time to answer the voice, and remain silent for the other time. Therefore, it is necessary to determine the response duration of the assessor within this response time period.

[0058] Exemplarily, obtain the response time period corresponding to each assessment item, and obtain the voice segment corresponding to this response time period from the assessment voice of each assessor, that is, intercept the voice segment corresponding to answering this assessment item. As Figure 5 shown, if the response time period is T1, then this voice segment can be intercepted from the assessment voice.

[0059] However, within this response time period, the assessor may not answer all at once. In other words, within this response time period, the assessor may answer multiple times. For example, for this assessment item, the assessor initially answered part of the content, paused for a while, and then answered part of the content. Therefore, within this response time period, the assessor may answer this assessment item once or multiple times. Therefore, perform voice feature recognition on this voice segment to obtain at least one first voice segment containing human voice in the first voice segment.

[0060] It should be noted that the voice assessment system of the present application supports the assessor to modify the answer. Therefore, within the response time period corresponding to each assessment item, in addition to answering the content corresponding to this assessment item, the assessor may also answer other content. For example, the assessor has answered this assessment item at the first moment, but when this assessment item has not ended yet, it is found that the answer is wrong, and the answer is immediately modified by voice answer. Since the present application does not calculate the duration of the modification, the duration used for this part of the answer should not be counted in this response duration; or, within the response time of this assessment item, it is recalled that the answer to the previous assessment item is wrong, and then the answer to the previous assessment item is immediately modified by voice answer within this response time period, and the time used for this part should also not be counted in the response duration.

[0061] Therefore, obtain a target first speech segment from at least one first speech segment, where the target first speech segment is the speech segment corresponding to each appraiser's first answer to each appraisal item. Exemplarily, perform text recognition on each first speech segment to obtain the first text content corresponding to each first speech segment; match each first text content with the appraisal item to obtain the similarity corresponding to each first text content. For example, the similarity can be determined through a machine reading comprehension model. For example, use the title of the appraisal item as the title in the machine reading comprehension model and the first text content as the answer, so as to determine the similarity between the appraisal item and each first text content based on the title and the answer; obtain the first text content with a similarity greater than the threshold, and use the first speech segment corresponding to the first text content where the similarity is greater than the threshold for the first time as the target first speech segment. Finally, use the total duration corresponding to the target first speech segment as the answering duration of each appraiser for each of the N appraisal items.

[0062] 403: Determine the answering content of each appraiser for each of the N appraisal items according to the appraisal speech of each appraiser.

[0063] It should be noted that during the appraisal process, appraisers may modify their answers. For example, after answering the first appraisal item, when answering the second appraisal item, the appraiser suddenly remembers that the answer to the first appraisal item is incorrect and needs to adjust the answering content of the first appraisal item. Then, the appraiser may adjust the answering content of the first appraisal item during the process of answering the second appraisal item. Therefore, in order to consider the situation where appraisers adjust the answers to appraisal items, after identifying at least one text content corresponding to each speech segment, it is necessary to adjust at least one text content corresponding to each speech segment to obtain the answering content for each appraisal item.

[0064] Exemplarily, perform text recognition on each first speech segment in at least one first speech segment to obtain the first text content corresponding to each first speech segment; according to the text content corresponding to each first speech segment, obtain at least one first text content corresponding to each of the N appraisal items. According to at least one first text content corresponding to each of the N appraisal items, determine the answering content of each appraiser for each of the N appraisal items.

[0065] Specifically, for the i-th evaluation item (i.e., any one of the N evaluation items), starting from the N-th evaluation item, traverse in reverse order, that is, traverse in the order from the N-th evaluation item, the (N - 1)-th evaluation item, …, the i-th evaluation item, and traverse the at least one first text content of each evaluation item traversed in reverse order; then, during the traversal, determine the first similarity between each first text content in the at least one first text content corresponding to each evaluation item traversed and the i-th evaluation item, where 1 ≤ i ≤ N and i is an integer value.

[0066] For example, as Figure 6 shown, when traversing the N-th evaluation item, first traverse in reverse order from the first text content corresponding to the last first speech segment of the N-th evaluation item, and calculate the first similarity corresponding to each text content in the at least one first text content of the N-th evaluation item in sequence.

[0067] After obtaining the first similarity corresponding to each traversal, determine whether the first similarity is greater than the first threshold. When the first similarity greater than the first threshold is first traversed, take the first text content corresponding to the first similarity first traversed as the answer content of the examiner for the i-th evaluation item, and the first text content in the remaining evaluation items will no longer be traversed; when traversing to the first first text content in the at least one first text content of the i-th evaluation item and all the first similarities obtained during the traversal are less than or equal to the first threshold, that is, no first similarity greater than the first threshold is found during the traversal, which means that the examiner has not answered the i-th evaluation item and has not made supplementary explanations for the i-th evaluation item when answering other evaluation items later. Therefore, take the empty set (Null) as the answer content of the examiner for the i-th evaluation item, that is, set the answer content of the i-th evaluation item to be blank.

[0068] 404: According to the answer content of each examiner for each evaluation item, determine the score and the number of modifications obtained by each examiner for each of the N evaluation items.

[0069] Exemplarily, according to the answer content of each examiner for each evaluation item and the standard answer of each evaluation item, determine the third similarity between the answer content of this evaluation item and the standard answer.

[0070] Specifically, obtain the question serial number of this evaluation item; according to the question serial number of this evaluation item, determine the question type of this evaluation item, where the question type includes objective questions and subjective questions. Then, according to the question type, answer content, and standard answer of this evaluation item, determine the third similarity of the evaluation item.

[0071] Exemplarily, when the question type of the evaluation item is an objective question, if the answer content of the evaluation item is the same as the standard answer, the third similarity corresponding to the evaluation item is determined to be 1; if the answer content of the evaluation item is different from the standard answer, the third similarity corresponding to the evaluation item is determined to be 0.

[0072] Exemplarily, when the question type of the evaluation item is a subjective question, keywords are extracted from the answer content corresponding to the first speech segment to obtain a first keyword set. Exemplarily, before extracting keywords from the answer content, abbreviations and short names in the answer content are mapped to obtain a new answer content, and then keywords are extracted from the new answer content to obtain a first keyword set. Since the keywords in the standard answer are all full names, mapping abbreviations and short names in the answer content in advance can improve the accuracy of determining the third similarity in case of incorrect keyword matching. Then, keywords are extracted from the standard answer corresponding to each evaluation item to obtain a second keyword set; the third similarity is determined according to the first keyword set and the second keyword set.

[0073] Exemplarily, the intersection of the first keyword set and the second keyword set is determined to obtain a third keyword set; then, the similarity between the i-th keyword and any one of the second remaining keywords is determined to obtain multiple second similarities corresponding to the i-th keyword, where the i-th keyword is any one of the first remaining keywords, the first remaining keywords are the remaining keywords in the first keyword set except the third keyword set, and the second remaining keywords are the remaining keywords in the second keyword set except the third keyword set; if the maximum second similarity among the multiple second similarities is greater than the threshold, the i-th keyword is used as a candidate keyword to obtain at least one candidate keyword in the first remaining keywords. Finally, a first ratio between the number of keywords in the third keyword set and the number of keywords in the second keyword set, and a second ratio between the number of keywords of at least one candidate keyword and the number of keywords in the second keyword set are obtained; the third similarity is determined according to the first ratio and the second ratio.

[0074] Exemplarily, the third similarity can be expressed by formula (1):

[0075] K1 = q1 + α * q2 Formula (1);

[0076] where K1 is the third similarity, q1 is the first ratio, q2 is the second ratio, and α is a preset weight coefficient, and α is greater than 0 and less than 1.

[0077] It should be noted that the reason for calculating the candidate keywords above is that there may be different expressions for the descriptions of some names. Therefore, as long as the response content of the appraiser is similar in essence, it can be regarded as a correct answer. However, the candidate keyword is a keyword that is relatively similar to a certain keyword in the second keyword set, but not exactly this keyword. Therefore, when using the candidate keyword to replace the keyword in the standard answer, a confidence discount needs to be given to it. Thus, a weight coefficient α is multiplied by q2, which discounts the confidence of the second ratio, that is, the second ratio is not fully trusted, while the first ratio is exactly the same keyword and its corresponding confidence does not need to be discounted. Therefore, the third similarity calculated by the above formula (1) has a relatively high accuracy.

[0078] Furthermore, after determining the third similarity corresponding to each appraisal item, the score obtained for each appraisal item can be determined based on the third similarity corresponding to each appraisal item and the preset score corresponding to each appraisal item. For example, the third similarity can be multiplied by the preset score to obtain the score obtained for each appraisal item.

[0079] Exemplarily, based on the response content of each appraiser for each appraisal item, determining the number of modification times of each appraiser for each appraisal item specifically includes:

[0080] For the i-th evaluation item, determine the second similarity between each first text content in at least one first text content corresponding to the j-th evaluation item and the i-th evaluation item, to obtain at least one second similarity, where the j-th evaluation item is any one of the N - i + 1 evaluation items from the i-th evaluation item to the N-th evaluation item, 1 ≤ i ≤ N, and the value of i is an integer. Similarly, the second similarity between each first text content and the i-th evaluation item can be determined by a machine reading comprehension model. Then, determine the number of second similarities greater than the second threshold in at least one second similarity corresponding to each of the N - i + 1 evaluation items, to obtain a first quantity corresponding to each of the N - i + 1 evaluation items, that is, for each evaluation item, determine the number of second similarities greater than the second threshold in at least one second similarity corresponding to each evaluation item, and use this number as the first quantity corresponding to this evaluation item. Therefore, N - i + 1 first quantities can be obtained for the N - i + 1 evaluation items. Then, use the sum of the N - i + 1 first quantities corresponding to the N - i + 1 evaluation items as the number of modifications of each appraiser for the i-th evaluation item, that is, sum the N - i + 1 first quantities, and use the summation result as the number of modifications of the i-th evaluation item; according to the number of modifications of each appraiser for the i-th evaluation item, obtain the number of modifications of each appraiser for each of the N evaluation items, that is, process each of the N evaluation items in the same way as the i-th evaluation item, and the number of modifications of each evaluation item can be obtained.

[0081] It can be seen that in the embodiment of the present application, when obtaining the scores of each evaluation item, the appraiser can dynamically adjust the answers to the evaluation items, improving the appraisal experience of the appraiser during the appraisal process. For each evaluation item, the present application uses the last answer content as the answer content of this evaluation item, so that the accuracy of the scores determined for each evaluation item is relatively high, and further the subsequent clustering accuracy is relatively high, realizing accurate information push.

[0082] Refer to Figure 7 , Figure 7 The functional unit composition block diagram of an information push device provided by an embodiment of the present application. The information push device 700 includes: an acquisition unit 701 and a processing unit 702;

[0083] The acquisition unit 701 is configured to acquire the appraisal situations of multiple appraisers for the pet doctor title;

[0084] The processing unit 702 is configured to determine the feature vector of each appraiser based on the appraisal situation of each appraiser among the multiple appraisers;

[0085] Cluster the multiple appraisers according to the feature vectors of each appraiser to obtain at least one clustering result;

[0086] Push the push information corresponding to each clustering result to the appraisers included in each clustering result among the at least one clustering result.

[0087] In some possible implementation manners, the pet doctor title includes N appraisal items, and the appraisal situation of each appraiser includes the answering duration, the number of modifications, and the obtained score of each appraiser for each of the N appraisal items; in terms of determining the feature vector of each appraiser based on the appraisal situation of each appraiser among the multiple appraisers, the processing unit 702 is specifically configured to:

[0088] Construct a first feature vector corresponding to each appraiser based on the answering duration of each appraiser for each of the N appraisal items;

[0089] Construct a second feature vector corresponding to each appraiser based on the number of modifications of each appraiser for each of the N appraisal items;

[0090] Construct a third feature vector corresponding to each appraiser based on the obtained score of each appraiser for each of the N appraisal items;

[0091] Concatenate the first feature vector, the second feature vector, and the third feature vector of each appraiser to obtain the feature vector of each appraiser.

[0092] In some possible implementation manners, before obtaining the appraisal situations of multiple appraisers for the pet doctor title, the obtaining unit 701 is further configured to obtain the appraisal voice of each appraiser among the multiple appraisers;

[0093] The processing unit 702 is further configured to determine the answering duration of each appraiser for each of the N appraisal items according to the appraisal voice of each appraiser;

[0094] Determine the answering content of each appraiser for each of the N appraisal items according to the appraisal voice of each appraiser;

[0095] Determine the obtained score and the number of modifications of each appraiser for each of the N appraisal items according to the answering content of each appraiser for each appraisal item.

[0096] In some possible embodiments, when determining, according to the evaluation voice of each evaluator, the response duration of each evaluator for each of the N evaluation items, the processing unit 702 is specifically configured to:

[0097] Obtain the response time period corresponding to each of the N evaluation items;

[0098] Obtain the voice segment corresponding to the response time period from the evaluation voice of each evaluator;

[0099] Perform voice feature detection on the voice segment to obtain at least one first voice segment containing human voices in the first voice segment;

[0100] Obtain a target first voice segment from the at least one first voice segment, where the target first voice segment is the voice segment corresponding to the first response of each evaluator to each of the N evaluation items;

[0101] Use the total duration of the target first voice segment as the response duration of each evaluator for each of the N evaluation items.

[0102] In some possible embodiments, when determining, according to the evaluation voice of each evaluator, the response content of each evaluator for each of the N evaluation items, the processing unit 702 is specifically configured to:

[0103] Perform text recognition on each of the at least one first voice segment to obtain the first text content corresponding to each of the first voice segments;

[0104] According to the text content corresponding to each of the first voice segments, obtain at least one first text content corresponding to each of the N evaluation items;

[0105] Determine the response content of each evaluator for each of the N evaluation items according to the at least one first text content corresponding to each of the N evaluation items.

[0106] In some possible embodiments, when determining, according to the at least one first text content corresponding to each of the N evaluation items, the response content of each evaluator for each of the N evaluation items, the processing unit 702 is specifically configured to:

[0107] For the i-th evaluation item, traverse in reverse order starting from the N-th evaluation item, and traverse in reverse order at least one first text content of each evaluation item during each traversal, and determine the first similarity between each first text content in the at least one first text content corresponding to each evaluation item during each traversal and the i-th evaluation item, where 1 ≤ i ≤ N and i is an integer;

[0108] When the first similarity greater than the first threshold is first traversed, the first text content corresponding to the first similarity is used as the answer content of the evaluator for the i-th evaluation item;

[0109] When the first first text content in at least one first text content of the i-th evaluation item is traversed and all the first similarities obtained during the traversal are less than or equal to the first threshold, an empty set is used as the answer content of the evaluator for the i-th evaluation item;

[0110] According to the answer content of the evaluator for the i-th evaluation item, determine the answer content of each evaluator for each of the N evaluation items.

[0111] In some possible implementation manners, in terms of determining the number of modifications of each evaluator for each evaluation item according to the answer content of each evaluator for each evaluation item, the processing unit 702 is specifically configured to:

[0112] For the i-th evaluation item, determine the second similarity between each first text content in the at least one first text content corresponding to the j-th evaluation item and the i-th evaluation item, and obtain at least one second similarity, where the j-th evaluation item is any one of the N - i + 1 evaluation items from the i-th evaluation item to the N-th evaluation item, 1 ≤ i ≤ N and i is an integer;

[0113] Determine the number of those greater than the second threshold among the at least one second similarity corresponding to each of the N - i + 1 evaluation items, and obtain the first quantity corresponding to each of the N - i + 1 evaluation items;

[0114] Take the sum of the N - i + 1 first quantities corresponding to the N - i + 1 evaluation items as the number of modifications of each evaluator for the i-th evaluation item;

[0115] According to the number of modifications of each evaluator for the i-th evaluation item, obtain the number of modifications of each evaluator for each of the N evaluation items.

[0116] Refer to Figure 8 ,Figure 8 This is a schematic structural diagram of an electronic device provided by an embodiment of the present application. As Figure 8 shown, the electronic device 800 includes a transceiver 801, a processor 802, and a memory 803. They are connected through a bus 804. The memory 803 is used to store computer programs and data, and can transmit the data stored in the memory 803 to the processor 802.

[0117] The processor 802 is used to read the computer program in the memory 803 and perform the following operations:

[0118] Obtain the evaluation situations of multiple evaluators for the title of pet doctor;

[0119] Based on the evaluation situation of each evaluator among the multiple evaluators, determine the feature vector of each evaluator;

[0120] Cluster the multiple evaluators according to the feature vector of each evaluator to obtain at least one clustering result;

[0121] Push the push information corresponding to each clustering result to the evaluators included in each clustering result among the at least one clustering result.

[0122] In some possible implementation manners, the title of pet doctor includes N evaluation items, and the evaluation situation of each evaluator includes the answering duration, the number of modifications, and the obtained score of each evaluator for each of the N evaluation items; in terms of determining the feature vector of each evaluator based on the evaluation situation of each evaluator among the multiple evaluators, the processor 802 is specifically used to perform the following operations:

[0123] Construct a first feature vector corresponding to each evaluator based on the answering duration of each evaluator for each of the N evaluation items;

[0124] Construct a second feature vector corresponding to each evaluator based on the number of modifications of each evaluator for each of the N evaluation items;

[0125] Construct a third feature vector corresponding to each evaluator based on the obtained score of each evaluator for each of the N evaluation items;

[0126] Concatenate the first feature vector, the second feature vector, and the third feature vector of each evaluator to obtain the feature vector of each evaluator.

[0127] In some possible implementation manners, before obtaining the evaluation situations of multiple evaluators for the title of pet doctor, the processor 802 is further used to perform the following operations:

[0128] Obtain the evaluation voice of each of the multiple evaluators;

[0129] According to the evaluation voice of each evaluator, determine the response duration of each evaluator for each of the N evaluation items;

[0130] According to the evaluation voice of each evaluator, determine the response content of each evaluator for each of the N evaluation items;

[0131] According to the response content of each evaluator for each evaluation item, determine the score and the number of modifications obtained by each evaluator for each of the N evaluation items.

[0132] In some possible implementation manners, in terms of determining the response duration of each evaluator for each of the N evaluation items according to the evaluation voice of each evaluator, the processor 802 is specifically configured to perform the following operations:

[0133] Obtain the response time period corresponding to each of the N evaluation items;

[0134] Obtain the voice segment corresponding to the response time period from the evaluation voice of each evaluator;

[0135] Perform voice feature detection on the voice segment to obtain at least one first voice segment containing human voices in the first voice segment;

[0136] Obtain a target first voice segment from the at least one first voice segment, where the target first voice segment is the voice segment corresponding to the first response of each evaluator to each of the N evaluation items;

[0137] Take the total duration corresponding to the target first voice segment as the response duration of each evaluator for each of the N evaluation items.

[0138] In some possible implementation manners, in terms of determining the response content of each evaluator for each of the N evaluation items according to the evaluation voice of each evaluator, the processor 802 is specifically configured to perform the following operations:

[0139] Perform text recognition on each of the at least one first voice segment to obtain first text content corresponding to each of the first voice segments;

[0140] According to the text content corresponding to each of the first voice segments, obtain at least one first text content corresponding to each of the N evaluation items;

[0141] Determine the response content of each appraiser for each of the N appraisal items based on at least one first text content corresponding to each of the N appraisal items.

[0142] In some possible implementation manners, in determining the response content of each appraiser for each of the N appraisal items based on at least one first text content corresponding to each of the N appraisal items, the processor 802 is specifically configured to perform the following operations:

[0143] For the i-th appraisal item, traverse in reverse order starting from the N-th appraisal item, and traverse in reverse order the at least one first text content of each traversed appraisal item, and determine the first similarity between each first text content in the at least one first text content corresponding to each traversed appraisal item and the i-th appraisal item, where 1 ≤ i ≤ N and i is an integer;

[0144] When the first similarity greater than the first threshold is first traversed, use the first text content corresponding to the first similarity as the response content of the appraiser for the i-th appraisal item;

[0145] When the first first text content in the at least one first text content of the i-th appraisal item is traversed and all the first similarities obtained during the traversal process are less than or equal to the first threshold, use the empty set as the response content of the appraiser for the i-th appraisal item;

[0146] Determine the response content of each appraiser for each of the N appraisal items based on the response content of the appraiser for the i-th appraisal item.

[0147] In some possible implementation manners, in determining the number of modification times of each appraiser for each appraisal item based on the response content of each appraiser for each appraisal item, the processor 802 is specifically configured to perform the following operations:

[0148] For the i-th appraisal item, determine the second similarity between each first text content in the at least one first text content corresponding to the j-th appraisal item and the i-th appraisal item, to obtain at least one second similarity, where the j-th appraisal item is any one of the N - i + 1 appraisal items from the i-th appraisal item to the N-th appraisal item, 1 ≤ i ≤ N and i is an integer;

[0149] Determine the number of at least one second similarity corresponding to each of the N - i + 1 evaluation items that is greater than the second threshold, and obtain a first quantity corresponding to each of the N - i + 1 evaluation items;

[0150] Take the sum of the N - i + 1 first quantities corresponding to the N - i + 1 evaluation items as the number of modifications made by each appraiser for the i-th evaluation item;

[0151] Based on the number of modifications made by each appraiser for the i-th evaluation item, obtain the number of modifications made by each appraiser for each of the N evaluation items.

[0152] Specifically, the above transceiver 801 may be Figure 7 the acquisition unit 701 of the information push device 700 in the above embodiment, and the above processor 802 may be Figure 7 the processing unit 702 of the information push device 700 in the above embodiment.

[0153] It should be understood that the electronic devices in this application may include smart phones (such as Android phones, iOS phones, Windows Phone phones, etc.), tablet computers, handheld computers, laptop computers, mobile Internet devices MID (Mobile Internet Devices, abbreviated as MID), or wearable devices, etc. The above are only examples of electronic devices and are not exhaustive, including but not limited to the above electronic devices. In practical applications, the above electronic devices may also include: intelligent vehicle terminals, computer devices, and so on.

[0154] This application embodiment also provides a computer-readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement some or all of the steps of any one of the information push methods described in the above method embodiments.

[0155] This application embodiment also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps of any one of the information push methods described in the above method embodiments.

[0156] It should be noted that, for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application.

[0157] In the above embodiments, the descriptions of the respective embodiments have their own focuses. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0158] In several embodiments provided by this application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical or other forms.

[0159] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0160] In addition, in each embodiment of this application, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware or in the form of software program modules.

[0161] When the integrated unit is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The aforementioned memory includes: various media such as USB flash drives, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), external hard drives, magnetic disks, or optical discs that can store program codes.

[0162] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing relevant hardware through a program. This program can be stored in a computer-readable memory, and the memory can include: flash drives, read-only memories (English: Read-Only Memory, abbreviated: ROM), random access memories (English: Random Access Memory, abbreviated: RAM), magnetic disks, or optical discs, etc.

[0163] The above has introduced the embodiments of this application in detail. Specific examples are used in this article to elaborate on the principle and implementation manner of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application; at the same time, for those of ordinary skill in the art, according to the idea of this application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to this application.

Claims

1. An information push method, characterized in that, The professional titles of pet veterinarians include N assessment items. The assessment situation of each assessor includes the answering duration, the number of modifications, and the obtained score of each assessor for each of the N assessment items. It includes: Obtain the assessment voices of each of multiple assessors; determine the answering duration of each assessor for each of the N assessment items according to the assessment voices of each assessor; determine the answering content of each assessor for each of the N assessment items according to the assessment voices of each assessor; determine the obtained score and the number of modifications of each assessor for each of the N assessment items according to the answering content of each assessor for each assessment item; The step of determining the answering duration of each assessor for each of the N assessment items according to the assessment voices of each assessor includes: Obtain the answering time period corresponding to each of the N assessment items; obtain the voice segment corresponding to the answering time period from the assessment voices of each assessor; perform human voice feature detection on the voice segment to obtain at least one first voice segment containing human voices in the voice segment; obtain the target first voice segment from the at least one first voice segment, where the target first voice segment is the voice segment corresponding to the first answer of each assessor for each of the N assessment items; use the total duration of the target first voice segment as the answering duration of each assessor for each of the N assessment items; The step of determining the answering content of each assessor for each of the N assessment items according to the assessment voices of each assessor includes: Perform text recognition on each of the at least one first speech segment to obtain first text content corresponding to each of the first speech segments; according to the text content corresponding to each of the first speech segments, obtain at least one first text content corresponding to each of the N assessment items; for the i-th assessment item, perform reverse traversal starting from the N-th assessment item, and perform reverse traversal on at least one first text content of the assessment item for each traversal, and determine the first similarity between each first text content in the at least one first text content corresponding to the assessment item for each traversal and the i-th assessment item, where 1 ≤ i ≤ N and i is an integer; when the first similarity greater than the first threshold is first traversed, then use the first text content corresponding to the first similarity as the answer content of the assessor for the i-th assessment item; when the first first text content in at least one first text content of the i-th assessment item is traversed and all the first similarities obtained during the traversal process are less than or equal to the first threshold, then use the empty set as the answer content of the assessor for the i-th assessment item; according to the answer content of the assessor for the i-th assessment item, determine the answer content of each assessor for each of the N assessment items; Obtain the assessment situations of multiple assessors for the title of veterinarian; Based on the assessment situation of each assessor among the multiple assessors, determine the feature vector of each assessor; Cluster the multiple assessors according to the feature vector of each assessor to obtain at least one clustering result; Push the push information corresponding to each clustering result to the assessors included in each clustering result among the at least one clustering result.

2. The method according to claim 1, characterized in that, The determining the feature vector of each assessor based on the assessment situation of each assessor among the multiple assessors includes: Construct a first feature vector corresponding to each assessor based on the answer duration of each assessor for each of the N assessment items; Construct a second feature vector corresponding to each assessor based on the number of modification times of each assessor for each of the N assessment items; Construct a third feature vector corresponding to each assessor based on the score obtained by each assessor for each of the N assessment items; Concatenate the first feature vector, the second feature vector, and the third feature vector of each assessor to obtain the feature vector of each assessor.

3. The method according to claim 1, characterized in that, According to the answer content of each assessor for each assessment item, determine the number of modification times of each assessor for each assessment item, including: For the i-th evaluation item, determine the second similarity between each of at least one first text content corresponding to the j-th evaluation item and the i-th evaluation item, to obtain at least one second similarity, where the j-th evaluation item is any one of the N - i + 1 evaluation items from the i-th evaluation item to the N-th evaluation item, 1 ≤ i ≤ N, and the value of i is an integer; Determine the number of second similarities greater than a second threshold among at least one second similarity corresponding to each of the N - i + 1 evaluation items, to obtain a first number corresponding to each of the N - i + 1 evaluation items; Take the sum of the N - i + 1 first numbers corresponding to the N - i + 1 evaluation items as the number of modifications made by each appraiser for the i-th evaluation item; Based on the number of modifications made by each appraiser for the i-th evaluation item, obtain the number of modifications made by each appraiser for each of the N evaluation items.

4. An information push device, characterized in that, The apparatus is used to implement the method according to any one of claims 1 - 3. The apparatus includes: an acquisition unit and a processing unit; The acquisition unit is used to acquire the evaluation situations of multiple appraisers for the title of veterinarian; The processing unit is used to determine the feature vector of each appraiser based on the evaluation situation of each appraiser among the multiple appraisers; Cluster the multiple appraisers according to the feature vector of each appraiser to obtain at least one clustering result; Push push information corresponding to each clustering result to the appraisers included in each clustering result among the at least one clustering result.

5. An electronic device, characterized in that, Including: A processor and a memory. The processor is connected to the memory. The memory is used to store a computer program. The processor is used to execute the computer program stored in the memory so that the electronic device executes the method according to any one of claims 1 - 3.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method according to any one of claims 1 - 3.

Images