Public praise data analysis method and device and electronic equipment

Abstract

The invention provides a public praise data analysis method and device and electronic equipment, relates to the technical field of automobile data mining, and is used for solving the problem of how to improve the accuracy of public praise scores. The method comprises the following steps: acquiring public opinion data to be analyzed and vehicle delivery data; wherein the public opinion data to be analyzed does not contain vehicle model information, and the vehicle delivery data comprises delivery information of at least one delivered vehicle model; performing data classification on the to-be-analyzed public opinion data to obtain an actual classification result of the to-be-analyzed public opinion data; and generating a public praise score of each delivered vehicle model based on the historical public opinion data, the vehicle delivery data and the actual classification result.

Description

Technical Field

[0001] The present disclosure relates to the technical field of automobile data mining, and in particular to a word-of-mouth data analysis method, device and electronic device. Background Art

[0002] In the prior art, in order to analyze product problems of sold models, automobile manufacturers usually collect automobile word-of-mouth data (such as purchase recommendations or user experience of corresponding models) published by users during the process of buying and using the car. Afterwards, the automobile word-of-mouth data is analyzed to obtain customer sentiment, accurately locate product problems, and thus meet customer needs.

[0003] At present, automobile manufacturers usually use manual labeling to label automobile reputation data. For example, for automobile reputation data with car models, the corresponding labeling results can be obtained based on the automobile reputation data, and the reputation score of the car model can be determined based on the labeling results, and relevant analysis can be performed based on the reputation score; for automobile reputation data without car models, although the labeling results corresponding to the automobile reputation data can be determined, since the automobile reputation data does not contain a car model, it is impossible to analyze the reputation score of the sold car model based on the automobile reputation data, and it is impossible to perform subsequent analysis work. Summary of the invention

[0004] In order to solve the above technical problems, the present disclosure provides a word-of-mouth data analysis method, device and electronic device.

[0005] In order to achieve the above objectives, the present invention adopts the following technical solutions:

[0006] In a first aspect, the present disclosure provides a word-of-mouth data analysis method, comprising: obtaining public opinion data to be analyzed and vehicle delivery data; wherein the public opinion data to be analyzed does not contain vehicle model information, and the vehicle delivery data includes delivery information of at least one delivered vehicle model; performing data classification on the public opinion data to be analyzed to obtain an actual classification result of the public opinion data to be analyzed; based on historical public opinion data, vehicle delivery data and actual classification results, generating a word-of-mouth score for each delivered vehicle model; wherein the historical public opinion data includes public opinion data that does not mention the delivered vehicle model and the actual classification result is the target classification result.

[0007] In some feasible examples, one historical public opinion data corresponds to one data source; based on the historical public opinion data, vehicle delivery data and actual classification results, a reputation score for each delivered vehicle model is generated, including: determining the delivery ratio of each delivered vehicle model based on the ratio of the delivery information of each delivered vehicle model to the sum of the delivery information; determining the first classification ratio of the target classification result based on the ratio of the total number of historical public opinion data that does not mention the delivered vehicle model and whose actual classification results are the target classification results to the total number of historical public opinion data; determining the second classification ratio of the target classification result based on the total number of historical classification results corresponding to the delivered vehicle model as the target classification results and the total number of historical classification results corresponding to the delivered vehicle model; performing the following operations on the historical public opinion data of the same data source: generating a theoretical score based on the delivery ratio, the first classification ratio and the second classification ratio; generating a theoretical score for the delivered vehicle model based on the theoretical score corresponding to each data source.

[0008] In some feasible examples, the following operations are performed on historical public opinion data from the same data source: based on the delivery ratio, the proportion of the first category and the proportion of the second category, a theoretical score is generated, including: the following operations are performed on historical public opinion data from the same data source: based on the product of the delivery ratio and the proportion of the first category, and the sum of the proportion of the second category, a reputation score for the delivered vehicle model is generated.

[0009] In some feasible examples, the word-of-mouth score satisfies the following formula:

[0010]

[0011] α+β=1;

[0012] Among them, w(v i ) represents the reputation score of the delivered model i, Indicates the delivery ratio, represents the proportion of the first category, Indicates the proportion of the second category, N(v i ) indicates the delivery information of the delivered models, N(v 总 ) represents the sum of the delivery information, N(nv i ) indicates the total number of historical public opinion data that do not mention the delivered models and whose actual classification results are the target classification results. N(wv 总 ) represents the total number of historical public opinion data, M(nv i ) indicates that the historical classification results corresponding to the delivered models are the total number of target classification results, M(wv i ) represents the historical classification results corresponding to the delivered models, and α and β are both constants greater than or equal to 0.

[0013] In some feasible examples, a theoretical score of a delivered vehicle model is generated based on the theoretical score corresponding to each data source, including: obtaining a weight score of the data source based on the product of the influence weight corresponding to the data source and the theoretical score corresponding to the data source; and obtaining the theoretical score of the delivered vehicle model based on the sum of the weight scores corresponding to each data source.

[0014] In some feasible examples, the word-of-mouth data analysis method provided by the present disclosure also includes: generating word-of-mouth analysis results based on the word-of-mouth score of each delivered vehicle model.

[0015] In the second aspect, the present disclosure provides a word-of-mouth data analysis device, including: an acquisition unit, used to acquire public opinion data to be analyzed and vehicle delivery data; wherein the public opinion data to be analyzed does not contain vehicle model information, and the vehicle delivery data includes delivery information of at least one delivered vehicle model; a processing unit, used to perform data classification on the public opinion data to be analyzed acquired by the acquisition unit, and obtain an actual classification result of the public opinion data to be analyzed; the processing unit is also used to generate a word-of-mouth score for each delivered vehicle model based on historical public opinion data, vehicle delivery data and actual classification results; wherein the historical public opinion data includes public opinion data that does not mention the delivered vehicle model and the actual classification result is the target classification result.

[0016] In some feasible examples, one historical public opinion data corresponds to one data source; the processing unit is specifically used to determine the delivery ratio of each delivered model based on the ratio of the delivery information of each delivered model acquired by the acquisition unit to the sum of the delivery information acquired by the acquisition unit; the processing unit is specifically used to determine the first classification ratio of the target classification result based on the ratio of the total number of historical public opinion data that do not mention the delivered model and whose actual classification results are the target classification results to the total number of historical public opinion data; the processing unit is specifically used to determine the second classification ratio of the target classification result based on the total number of historical classification results corresponding to the delivered model as the target classification result and the total number of historical classification results corresponding to the delivered model; the processing unit is specifically used to perform the following operations on the historical public opinion data of the same data source: generate a theoretical score based on the delivery ratio, the first classification ratio and the second classification ratio; the processing unit is specifically used to generate a theoretical score for the delivered model based on the theoretical score corresponding to each data source.

[0017] In some feasible examples, the processing unit is specifically used to perform the following operations on the historical public opinion data of the same data source: based on the product of the delivery ratio and the proportion of the first category, and the sum of the proportion of the second category, generate a reputation score for the delivered vehicle model.

[0018] In some feasible examples, the word-of-mouth score satisfies the following formula:

[0019]

[0020] α+β=1;

[0021] Among them, w(v i ) represents the reputation score of the delivered model i, Indicates the delivery ratio, represents the proportion of the first category, Indicates the proportion of the second category, N(v i ) indicates the delivery information of the delivered models, N(v 总 ) represents the sum of the delivery information, N(nv i ) indicates the total number of historical public opinion data that do not mention the delivered models and whose actual classification results are the target classification results. N(wv 总 ) represents the total number of historical public opinion data, M(nv i ) indicates that the historical classification results corresponding to the delivered models are the total number of target classification results, M(wv i ) represents the historical classification results corresponding to the delivered models, and α and β are both constants greater than or equal to 0.

[0022] In some feasible examples, the processing unit is specifically used to obtain a weight score of a data source based on the product of an influence weight corresponding to the data source and a theoretical score corresponding to the data source; the processing unit is specifically used to obtain a theoretical score of a delivered vehicle model based on the sum of the weight scores corresponding to each data source.

[0023] In some feasible examples, the processing unit is further used to generate word-of-mouth analysis results based on the word-of-mouth score of each delivered vehicle model.

[0024] In a third aspect, the present disclosure provides an electronic device, comprising: a memory and a processor, the memory being used to store a computer program; the processor being used to enable the electronic device to implement the word-of-mouth data analysis method provided in the first aspect when executing the computer program.

[0025] In a fourth aspect, the present disclosure provides a computer-readable storage medium, comprising: a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a computing device, the computing device implements the word-of-mouth data analysis method provided in the first aspect above.

[0026] In a fifth aspect, the present disclosure provides a vehicle, comprising any word-of-mouth data analysis device provided in the second aspect.

[0027] In the present disclosure, the name of the above-mentioned word-of-mouth data analysis device does not limit the device or functional module itself. In actual implementation, these devices or functional modules may appear with other names. As long as the functions of each device or functional module are similar to those of the present disclosure, they fall within the scope of the claims of the present disclosure and their equivalent technologies.

[0028] These and other aspects of the present disclosure will become more apparent from the following description.

[0029] Compared with the prior art, the technical solution provided by the present invention has the following advantages:

[0030] The word-of-mouth data analysis method provided by the present disclosure, when analyzing the word-of-mouth score, by introducing public opinion data that does not mention the delivered models and whose actual classification results are the target classification results, the amount of data for calculating the word-of-mouth score is greatly increased. At the same time, by introducing vehicle delivery data, the word-of-mouth score of each model can be better located, such as: by obtaining the public opinion data to be analyzed and the vehicle delivery data, since the public opinion data to be analyzed does not contain vehicle model information, it is necessary to classify the public opinion data to be analyzed to obtain the actual classification results of the public opinion data to be analyzed. Afterwards, based on the historical public opinion data, vehicle delivery data and actual classification results, the word-of-mouth score of each delivered model is generated. Since the amount of data for calculating the word-of-mouth score is greatly increased, the representation of the word-of-mouth score is more comprehensive, so the accuracy of the word-of-mouth score can be improved, solving the problem of how to improve the accuracy of the word-of-mouth score. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0033] Figure 1 One of the flowcharts of the word-of-mouth data analysis method provided in the embodiment of the present disclosure;

[0034] Figure 2 A schematic diagram of a scene of a first vehicle and a target vehicle in the word-of-mouth data analysis method provided in an embodiment of the present disclosure;

[0035] Figure 3 The second flowchart of the word-of-mouth data analysis method provided by the embodiment of the present disclosure;

[0036] Figure 4 The third flowchart of the word-of-mouth data analysis method provided by the embodiment of the present disclosure;

[0037] Figure 5 A fourth flowchart of the word-of-mouth data analysis method provided in an embodiment of the present disclosure;

[0038] Figure 6 A schematic diagram of the structure of a word-of-mouth data analysis device provided in an embodiment of the present disclosure;

[0039] Figure 7 A schematic diagram of the structure of a server provided in an embodiment of the present disclosure;

[0040] Figure 8 A schematic diagram of the structure of a computer program product provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0041] In order to more clearly understand the above-mentioned objectives, features and advantages of the present disclosure, the scheme of the present disclosure will be further described below. It should be noted that the embodiments of the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0042] In the following description, many specific details are set forth to facilitate a full understanding of the present disclosure, but the present disclosure may also be implemented in other ways different from those described herein; it is obvious that the embodiments in the specification are only part of the embodiments of the present disclosure, rather than all of the embodiments.

[0043] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0044] GPT in the embodiments of the present disclosure refers to Generative Pre-Trained Transformer (generative pre-trained Transformer model), which is an Internet-based deep learning model for text generation that can be trained with available data.

[0045] Ada in the disclosed embodiment is the latest model launched by OpenAI. It is a large-scale, multi-task language model that can perform a variety of natural language tasks, such as question answering, generation, classification, etc.

[0046] Davinci in the disclosed embodiment is the largest language model launched by OpenAI. It consists of 17.5 billion parameters and is one of the most advanced and powerful language models currently available.

[0047] Babbage in the disclosed embodiment is a medium-sized language model launched by OpenAI, which consists of 600 million parameters and can be used for generating text, answering questions and classification tasks, etc.

[0048] Curie in the disclosed embodiment is a medium-to-large language model launched by OpenAI. It consists of 1.3 billion parameters and can be used for tasks such as natural language generation, question answering and text classification.

[0049] The executor of the word-of-mouth data analysis method in the embodiment of the present application is a word-of-mouth data analysis device, which can be set in a server or other equipment, and this embodiment does not limit this.

[0050] The delivery information involved in the present disclosure may be data authorized by the user or fully authorized by all parties.

[0051] Exemplarily, the word-of-mouth data analysis method provided by the embodiment of the present disclosure is introduced by taking the execution subject of the word-of-mouth data analysis method provided by the embodiment of the present disclosure as a server as an example.

[0052] like Figure 1 As shown, the word-of-mouth data analysis method provided by the embodiment of the present disclosure includes the following steps S11-S13:

[0053] S11, obtaining public opinion data to be analyzed and vehicle delivery data, wherein the public opinion data to be analyzed does not include vehicle model information, and the vehicle delivery data includes delivery information of at least one delivered vehicle model.

[0054] In some examples, the automobile manufacturer obtains the delivery information of the delivered models by collecting the sales records of the delivered models, wherein the delivery information includes the delivery quantity.

[0055] In some examples, the public opinion data to be analyzed may be posts uploaded by users in a free forum of an automobile manufacturer, or questions raised by users, or may be obtained from other websites where users communicate about vehicle-related issues.

[0056] S12. Classify the public opinion data to be analyzed to obtain actual classification results of the public opinion data to be analyzed.

[0057] In some examples, the public opinion data to be analyzed can be input into a pre-trained sentiment classification model for data classification to obtain the actual classification results of the public opinion data to be analyzed. The training process of the pre-trained sentiment classification model is as follows:

[0058] The training sample data and the labeling results of the training sample data are obtained. The training sample data includes different car models and car reputation data of car manufacturers obtained on the self-owned forums and vertical websites in the automotive field, and the labeling results include the classification results of the car reputation data.

[0059] Exemplarily, the training sample data includes historical public opinion data on specific car models and historical public opinion data that only mentions car manufacturers but not specific car models. When labeling the training sample data, a preset number (e.g., 20,000) of historical public opinion data can be labeled, and the ratio of historical public opinion data on specific car models to historical public opinion data that only mentions car manufacturers but not specific car models is a preset ratio, such as 3:7.

[0060] Based on the training sample data and the labeled results of the training sample data, the deep learning model (such as the Davinci model under the GPT framework, the Curie model under the GPT framework, the Babbage model under the GPT framework, or any one of the Ada models under the GPT framework) is fine-tuned (prediction accuracy) to obtain a sentiment classification model.

[0061] Exemplarily, the actual classification result includes any one of negative, positive, and neutral.

[0062] In some examples, different actual classification results correspond to different classification keywords. When classifying the public opinion data to be analyzed, the public opinion data to be analyzed is first segmented to obtain at least one actual segmented word. Then, the similarity between the actual segmented word and the classification keyword is calculated (such as cosine similarity or Euclidean distance). Then, the actual classification result to which the classification keyword corresponding to the maximum similarity belongs is taken as the actual classification result of the public opinion data to be analyzed.

[0063] S13. Generate a reputation score for each delivered vehicle model based on historical public opinion data, vehicle delivery data, and actual classification results. The historical public opinion data includes public opinion data that does not mention the delivered vehicle model and whose actual classification results are the target classification results.

[0064] In some examples, one historical public opinion data corresponds to one data source. When generating the reputation score of each delivered model based on the historical public opinion data, vehicle delivery data and actual classification results, the delivery ratio of each delivered model can be determined based on the ratio of the delivery information (such as the delivery quantity) of each delivered model to the sum of the delivery information (such as the total delivery quantity of all models). Afterwards, the historical public opinion data of the same data source are subjected to the following operations: based on the ratio of the total number of historical public opinion data that do not mention the delivered model and the actual classification result is the target classification result to the total number of historical public opinion data, the first classification ratio of the target classification result is determined. Based on the total number of historical classification results corresponding to the delivered model as the target classification result and the total number of historical classification results corresponding to the delivered model, the second classification ratio of the target classification result is determined. Afterwards, the historical public opinion data of the same data source are subjected to the following operations: based on the delivery ratio, the first classification ratio and the second classification ratio, a theoretical score is generated. For example, the reputation score of the delivered model is generated based on the product of the delivery ratio and the first classification ratio and the sum of the second classification ratio. Afterwards, the theoretical score of the delivered model is generated based on the theoretical score corresponding to each data source. For example, the weight score of the data source is obtained by multiplying the impact weight corresponding to the data source and the theoretical score corresponding to the data source. The theoretical score of the delivered model is obtained by summing the weight scores corresponding to each data source.

[0065] For example, as shown in Table 1, the data source includes the automotive field and pan-media, where the automotive field includes media 1 and media 2, the influence weight of media 1 is 0.8, the influence weight of media 2 is 0.7, and the pan-media includes media 3 and media 4, the influence weight of media 3 is 0.3, and the influence weight of media 4 is 0.2. The calculation process of the theoretical score of the delivered models is as follows:

[0066] Table 1

[0067]

[0068] Assume that the theoretical score of model 1 in media 1 is 0.5, the theoretical score of model 1 in media 2 is 0.3, the theoretical score of model 1 in media 3 is 0.9, and the theoretical score of model 1 in media 4 is 0.6. From the above, we can know that the theoretical score of model 1 is 0.5×0.5+0.3×0.7+0.9×0.3+0.6×0.2, which is equal to 0.85. In this way, the theoretical score of model 1 is obtained.

[0069] In some examples, word-of-mouth analysis results may be generated based on the theoretical score of each model. For example, a chart may be drawn according to the theoretical score of each model, and a target icon including the theoretical score of each model may be generated, so that the theoretical score of each model may be more intuitively displayed.

[0070] From the above, it can be seen that the word-of-mouth data analysis method provided by the present disclosure, when analyzing the word-of-mouth score, by introducing public opinion data that does not mention the delivered models and whose actual classification results are the target classification results, the amount of data for calculating the word-of-mouth score is greatly increased.

[0071] In some feasible examples, one historical public opinion data corresponds to one data source; combined with Figure 1 ,like Figure 2 As shown, the above S13 can be specifically implemented through the following S130-S134.

[0072] S130. Determine the delivery ratio of each delivered model based on the ratio of the delivery information of each delivered model to the sum of the delivery information.

[0073] For example, taking the delivery information as the delivery quantity, assume that the car manufacturer A has currently delivered models including model 1, model 2 and model 3. Among them, the delivery quantity of model 1 is 100, the delivery quantity of model 2 is 150, and the delivery quantity of model 3 is 50. At this time, the delivery ratio of model 1 is the ratio of the delivery quantity of model 1 100 to the sum of the delivery quantities (100+150+50), that is, The delivery ratio of model 2 is the ratio of the delivery quantity of model 2 (150) to the sum of the delivery quantities (100+150+50), that is, The delivery ratio of model 3 is the ratio of the delivery quantity of model 3, 50, to the sum of the delivery quantities (100+150+50), that is,

[0074] S131. Determine the first classification proportion of the target classification result based on the ratio of the total number of historical public opinion data that do not mention the delivered models and whose actual classification results are the target classification results to the total number of historical public opinion data.

[0075] In some examples, historical public opinion data of vehicle models delivered by automobile manufacturer A can be collected from different data sources. Therefore, it is necessary to calculate the first category proportion of each delivered vehicle model for each data source separately.

[0076] For example, combined with the example given in S13 above, assuming that Model 1 does not mention the delivered model (such as Model 1) in Media 1, and the total number of historical public opinion data with the actual classification result (such as negative) as the target classification result is 100, and the total number of historical public opinion data collected by Media 1 is 300, then the proportion of the first classification is

[0077] S132. Determine the second classification ratio of the target classification results based on the total number of historical classification results corresponding to the delivered models as the target classification results and the total number of historical classification results corresponding to the delivered models.

[0078] In some examples, historical public opinion data of vehicle models delivered by automobile manufacturer A can be collected from different data sources. By classifying the historical public opinion data, historical public opinion data containing delivered vehicle models (such as vehicle model 1) can be obtained. Afterwards, by classifying the historical public opinion data containing vehicle model 1, the data classification results corresponding to the historical public opinion data containing vehicle model 1 can be obtained. Afterwards, by counting the data classification results corresponding to the historical public opinion data of vehicle model 1, the total number of historical classification results corresponding to vehicle model 1 as target classification results (such as negative) and the total number of historical classification results corresponding to vehicle model 1 can be obtained. Furthermore, the second classification proportion of the target classification results can be determined based on the total number of historical classification results corresponding to delivered vehicle models as target classification results and the total number of historical classification results corresponding to delivered vehicle models.

[0079] S133. Perform the following operations on the historical public opinion data of the same data source: generate a theoretical score based on the delivery ratio, the first category ratio, and the second category ratio.

[0080] S134. Generate a theoretical score for the delivered vehicle model based on the theoretical score corresponding to each data source.

[0081] In some possible implementation examples, combined with Figure 2 ,like Figure 3 As shown, the above S133 can be specifically implemented through the following S1330.

[0082] S1330. Perform the following operation on the historical public opinion data of the same data source: Generate a reputation score for the delivered vehicle model based on the product of the delivery ratio and the first category proportion and the sum of the second category proportion.

[0083] In some feasible examples, the word-of-mouth score satisfies the following formula:

[0084]

[0085] α+β=1.

[0086] Among them, w(v i ) represents the reputation score of the delivered model i, Indicates the delivery ratio, represents the proportion of the first category, Indicates the proportion of the second category, N(v i ) indicates the delivery information of the delivered models, N(v 总 ) represents the sum of the delivery information, N(nv i ) indicates the total number of historical public opinion data that do not mention the delivered models and whose actual classification results are the target classification results. N(wv 总) represents the total number of historical public opinion data, M(nv i ) indicates that the historical classification results corresponding to the delivered models are the total number of target classification results, M(wv i ) represents the historical classification results corresponding to the delivered models, and α and β are both constants greater than or equal to 0.

[0087] In some examples, in order to ensure the accuracy of the word-of-mouth score, the word-of-mouth data analysis method provided in the embodiments of the present disclosure uses an AB comparison experiment with a one-month experimental period. A preset number (such as 100 groups) of historical public opinion data of vehicle type 1 are selected as a training set, and different values ​​are assigned to α and β respectively to verify the appropriate values ​​of α and β.

[0088] Exemplarily, in combination with the example given in S130 above, a preset number (such as 100 groups) of historical public opinion data of vehicle type 1 is selected as a training set, and three different sets of values ​​for α and β are set as an example. The values ​​of α and β are shown in Table 2.

[0089] Table 2

[0090] AB comparison experiment α β Accuracy Experiment 1 0.5 0.5 0.47 Experiment 2 0.6 0.4 0.57 Experiment 3 0.3 0.7 0.89

[0091] In this way, we can get the reputation score of 0.47 for Model 1 in Experiment 1, and use this reputation score as the accuracy rate. Similarly, we can get the reputation score of 0.57 for Model 1 in Experiment 2, and use this reputation score as the accuracy rate. Similarly, we can get the reputation score of 0.89 for Model 1 in Experiment 3, and use this reputation score as the accuracy rate.

[0092] Afterwards, by comparing the reputation scores of Model 1 under different experiments, the values ​​of α and β corresponding to the maximum reputation score are used as the values ​​of α and β in the reputation score corresponding to Model 1, that is, the value of α in the reputation score of Model 1 is 0.3 and the value of is 0.7. Similarly, the values ​​of α and β corresponding to Model 2 and Model 3 can be obtained respectively.

[0093] In some examples, the values ​​of α and β in the reputation scores of different car models may be the same or different, and are not limited here.

[0094] In some possible implementation examples, combined with Figure 2 ,like Figure 4 As shown, the above S134 can be specifically implemented through the following S1340 and S1341.

[0095] S1340. Obtain a weight score for the data source based on the product of the influence weight corresponding to the data source and the theoretical score corresponding to the data source.

[0096] S1341. Obtain a theoretical score for the delivered vehicle model based on the sum of the weighted scores corresponding to each data source.

[0097] In some possible implementation examples, combined with Figure 1 ,like Figure 5 As shown, the word-of-mouth data analysis method provided by the embodiment of the present disclosure also includes: S14.

[0098] S14. Generate word-of-mouth analysis results based on the word-of-mouth score of each delivered vehicle model.

[0099] In some examples, in order to prevent negative public opinion about the delivered models due to the polarization of the word-of-mouth scores of the delivered models, the word-of-mouth data analysis method provided in the embodiments of the present disclosure can generate a word-of-mouth analysis result based on the current word-of-mouth score of the delivered model and the historical word-of-mouth score of the delivered model, such as: based on the relationship between the reputation difference between the current word-of-mouth score and the historical word-of-mouth score and the warning threshold, generate a word-of-mouth analysis result, such as when the reputation difference is greater than the warning threshold, determine that the word-of-mouth analysis result of the delivered model is out of control public opinion; or, when the reputation difference is less than or equal to the warning threshold, determine that the word-of-mouth analysis result of the delivered model is controllable public opinion.

[0100] In some examples, the word-of-mouth data analysis method provided by the embodiments of the present disclosure pre-configures the correspondence between different word-of-mouth scores and word-of-mouth analysis results. Afterwards, the correspondence can be queried based on the word-of-mouth score of the delivered model to obtain the word-of-mouth analysis result of the delivered model.

[0101] Exemplarily, the corresponding relationship between the word-of-mouth score and the word-of-mouth analysis result is shown in Table 3.

[0102] Table 3

[0103] Score range Word of mouth analysis results [0,0.2) Poor reputation [0.2,0.6) Average reputation [0.6,0.8) Good reputation [0.8,1.0] Excellent reputation

[0104] For example, when the reputation score of the delivered model is 0.3, it can be known from querying Table 3 that the result of the reputation analysis is average. Therefore, it can be determined that the result of the reputation analysis of the delivered model is average.

[0105] Alternatively, when the word-of-mouth score of the delivered model is 0.9, it can be seen from querying Table 3 that the word-of-mouth analysis result is an excellent reputation. Therefore, it can be determined that: the word-of-mouth analysis result of the delivered model is an excellent reputation. The above mainly introduces the solution provided by the embodiment of the present disclosure from the perspective of method. In order to achieve the above functions, it includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiment disclosed in this article, the present disclosure can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present disclosure.

[0106] The embodiments of the present disclosure can divide the functional modules of the word-of-mouth data analysis device according to the above method examples. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in the embodiments of the present disclosure is schematic and is only a logical functional division. There may be other division methods in actual implementation.

[0107] like Figure 6 , which is a schematic diagram of the structure of a word-of-mouth data analysis device 10 provided in an embodiment of the present disclosure, includes an acquisition unit 101 and a processing unit 102 .

[0108] The acquisition unit 101 is used to acquire the public opinion data to be analyzed and the vehicle delivery data; wherein the public opinion data to be analyzed does not include vehicle model information, and the vehicle delivery data includes the delivery information of at least one delivered vehicle model; the processing unit 102 is used to perform data classification on the public opinion data to be analyzed acquired by the acquisition unit 101 to obtain the actual classification result of the public opinion data to be analyzed; the processing unit 102 is also used to generate a reputation score for each delivered vehicle model based on the historical public opinion data, the vehicle delivery data and the actual classification result; wherein the historical public opinion data includes public opinion data that does not mention the delivered vehicle model and the actual classification result is the target classification result.

[0109] In some feasible examples, one historical public opinion data corresponds to one data source; the processing unit 102 is specifically used to determine the delivery ratio of each delivered model based on the ratio of the delivery information of each delivered model obtained by the acquisition unit 101 to the sum of the delivery information obtained by the acquisition unit 101; the processing unit 102 is specifically used to determine the first classification ratio of the target classification result based on the ratio of the total number of historical public opinion data that do not mention the delivered model and whose actual classification results are the target classification results to the total number of historical public opinion data; the processing unit 102 is specifically used to determine the second classification ratio of the target classification result based on the total number of historical classification results corresponding to the delivered model being the target classification results and the total number of historical classification results corresponding to the delivered model; the processing unit 102 is specifically used to perform the following operations on the historical public opinion data of the same data source: generate a theoretical score based on the delivery ratio, the first classification ratio and the second classification ratio; the processing unit 102 is specifically used to generate a theoretical score for the delivered model based on the theoretical score corresponding to each data source.

[0110] In some feasible examples, the processing unit 102 is specifically used to perform the following operations on the historical public opinion data of the same data source: based on the product of the delivery ratio and the proportion of the first category, and the sum of the proportion of the second category, generate a reputation score for the delivered vehicle model.

[0111] In some feasible examples, the word-of-mouth score satisfies the following formula:

[0112]

[0113] α+β=1;

[0114] Among them, w(v i ) represents the reputation score of the delivered model i, Indicates the delivery ratio, represents the proportion of the first category, Indicates the proportion of the second category, N(v i ) indicates the delivery information of the delivered models, N(v 总 ) represents the sum of the delivery information, N(nv i ) indicates the total number of historical public opinion data that do not mention the delivered models and whose actual classification results are the target classification results. N(wv 总 ) represents the total number of historical public opinion data, M(nv i ) indicates that the historical classification results corresponding to the delivered models are the total number of target classification results, M(wv i ) represents the historical classification results corresponding to the delivered models, and α and β are both constants greater than or equal to 0.

[0115] In some implementable examples, the processing unit 102 is specifically used to obtain a weight score of a data source based on the product of an influence weight corresponding to the data source and a theoretical score corresponding to the data source; the processing unit 102 is specifically used to obtain a theoretical score of a delivered vehicle model based on the sum of the weight scores corresponding to each data source.

[0116] In some practicable examples, the processing unit 102 is further configured to generate a word-of-mouth analysis result based on the word-of-mouth score of each delivered vehicle model.

[0117] Among them, all relevant contents of each step involved in the above method embodiment can be referred to the functional description of the corresponding functional module, and its role will not be repeated here.

[0118] Of course, the word-of-mouth data analysis device 10 provided in the embodiment of the present disclosure includes but is not limited to the above modules, for example, the word-of-mouth data analysis device 10 may also include a storage unit 103. The storage unit 103 may be used to store the program code of the word-of-mouth data analysis device 10, and may also be used to store data generated by the word-of-mouth data analysis device 10 during operation, such as data in a write request, etc.

[0119] Figure 7 A schematic diagram of the structure of a server provided in an embodiment of the present disclosure is shown in FIG. Figure 7 As shown, the server may include: at least one processor 51 , a memory 52 , a communication interface 53 and a communication bus 54 .

[0120] Combine the following Figure 7 A detailed introduction to each component of the server:

[0121] The processor 51 is the control center of the electronic device 10, and may be a processor or a general term for multiple processing elements. For example, the processor 51 is a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present disclosure, such as one or more DSPs, or one or more field programmable gate arrays (FPGAs).

[0122] In a specific implementation, as an embodiment, the processor 51 may include one or more CPUs, such as Figure 7 Also, as an embodiment, the electronic device may include multiple processors, such as Figure 751 and 55 are shown in FIG. Each of these processors may be a single-core processor (Single-CPU) or a multi-core processor (Multi-CPU). The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0123] The memory 52 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 52 may exist independently and be connected to the processor 51 via a communication bus 54. The memory 52 may also be integrated with the processor 51.

[0124] In a specific implementation, the memory 52 is used to store the data in the present disclosure and execute the software program in the present disclosure. The processor 51 can execute various functions of the air conditioner by running or executing the software program stored in the memory 52 and calling the data stored in the memory 52.

[0125] The communication interface 53 uses any transceiver-like device to communicate with other devices or communication networks, such as Radio Access Network (RAN), Wireless Local Area Networks (WLAN), terminals, the cloud, etc. The communication interface 53 may include an acquisition unit to implement the acquisition function.

[0126] The communication bus 54 may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 7 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0127] As an example, combine Figure 6 The function of the acquisition unit 101 in the word-of-mouth data analysis device 10 is similar to Figure 7 The function of the communication interface 53 in the word-of-mouth data analysis device 10 is the same as that of the processing unit 102 in the word-of-mouth data analysis device 10. Figure 7 The function of the processor 51 in the word-of-mouth data analysis device 10 is the same as that of the storage unit 103 in the word-of-mouth data analysis device 10. Figure 7 The function of the memory 52 in is the same.

[0128] Another embodiment of the present disclosure further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a computing device, the computing device executes the word-of-mouth data analysis method shown in the above method embodiment.

[0129] In some embodiments, the disclosed methods may be implemented as computer program instructions encoded in a machine-readable format on a computer-readable storage medium or on other non-transitory media or articles of manufacture.

[0130] Figure 8 A conceptual partial view of a computer program product provided by an embodiment of the present disclosure is schematically shown, wherein the computer program product includes a computer program for executing a computer process on a computing device.

[0131] In one embodiment, the computer program product is provided using a signal bearing medium 410. The signal bearing medium 410 may include one or more program instructions, which when executed by one or more processors may provide the above-mentioned Figure 1 Thus, for example, reference to Figure 1 In the embodiment shown in , one or more features of S11-S13 can be undertaken by one or more instructions associated with the signal bearing medium 410. In addition, Figure 8 The program instructions in also describe example instructions.

[0132] In some examples, signal bearing medium 410 may include computer readable medium 411, such as, but not limited to, a hard drive, a compact disk (CD), a digital video disk (DVD), a digital tape, a memory, a read-only memory (ROM), or a random access memory (RAM), and the like.

[0133] In some implementations, signal bearing medium 410 may include computer recordable medium 412 such as, but not limited to, memory, read / write (R / W) CD, R / W DVD, and the like.

[0134] In some implementations, signal bearing medium 410 may include communication medium 413 such as, but not limited to, digital and / or analog communication media (eg, fiber optic cables, waveguides, wired communication links, wireless communication links, etc.).

[0135] Signal bearing medium 410 may be communicated by a wireless form of communication medium 413 (eg, wireless communication medium complying with IEEE 802.41 standard or other transmission protocol). The one or more program instructions may be, for example, computing device executable instructions or logic implementing instructions.

[0136] In some examples, such as for Figure 6 The described word-of-mouth data analysis device 10 may be configured to provide various operations, functions, or actions in response to one or more program instructions via the computer-readable medium 411 , the computer-recordable medium 412 , and / or the communication medium 413 .

[0137] Through the description of the above implementation methods, technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0138] In the several embodiments provided in the present disclosure, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0139] The units described as separate components may or may not be physically separated, and the components shown as units may be one physical unit or multiple physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0140] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0141] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present disclosure is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a device (which can be a single-chip microcomputer, chip, etc.) or a processor (processor) to perform all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard drives, ROM, RAM, magnetic disks or optical disks.

[0142] The above description is only a specific embodiment of the present disclosure, so that those skilled in the art can understand or implement the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A word-of-mouth data analysis method, characterized in that: include: Obtaining public opinion data to be analyzed and vehicle delivery data; wherein the public opinion data to be analyzed does not include vehicle model information, and the vehicle delivery data includes delivery information of at least one delivered vehicle model; Performing data classification on the public opinion data to be analyzed to obtain actual classification results of the public opinion data to be analyzed; Based on the historical public opinion data, the vehicle delivery data and the actual classification results, a reputation score for each of the delivered vehicle models is generated; wherein the historical public opinion data includes public opinion data that does not mention the delivered vehicle model and whose actual classification results are target classification results.

2. The word-of-mouth data analysis method according to claim 1, characterized in that: One historical public opinion data corresponds to one data source; The generating of the reputation score of each delivered vehicle model based on the historical public opinion data, the vehicle delivery data and the actual classification result includes: Determining a delivery ratio of each of the delivered models based on a ratio of the delivery information of each of the delivered models to the sum of the delivery information; Determine the first classification proportion of the target classification result based on the ratio of the total number of historical public opinion data that do not mention the delivered model and whose actual classification results are the target classification results to the total number of the historical public opinion data; Based on the total number of the historical classification results corresponding to the delivered models being the target classification results, and the total number of the historical classification results corresponding to the delivered models, determining the second classification proportion of the target classification results; For the historical public opinion data from the same data source, the following operations are performed: generating a theoretical score based on the delivery ratio, the first category ratio, and the second category ratio; Based on the theoretical score corresponding to each data source, a theoretical score of the delivered vehicle model is generated.

3. The word-of-mouth data analysis method according to claim 2, characterized in that: The following operations are performed on the historical public opinion data of the same data source: generating a theoretical score based on the delivery ratio, the first classification ratio and the second classification ratio, including: The following operation is performed on the historical public opinion data of the same data source: based on the product of the delivery ratio and the proportion of the first category, and the sum of the proportion of the second category, a reputation score for the delivered vehicle model is generated.

4. The word-of-mouth data analysis method according to claim 3, characterized in that: The word-of-mouth score satisfies the following formula: α+β＝1； Among them, w(v i ) represents the reputation score of the delivered model i, Indicates the delivery ratio, represents the proportion of the first category, Indicates the proportion of the second category, N(v i ) represents the delivery information of the delivered model, N(v 总 ) represents the sum of the delivery information, N(nv i ) indicates the total number of historical public opinion data that do not mention the delivered models and whose actual classification results are the target classification results. N(wv 总 ) represents the total number of historical public opinion data, M(nv i ) indicates that the historical classification results corresponding to the delivered models are the total number of the target classification results, M(wv i ) represents the historical classification results corresponding to the delivered models, and α and β are both constants greater than or equal to 0.

5. The word-of-mouth data analysis method according to claim 2, characterized in that: The generating the theoretical score of the delivered vehicle model based on the theoretical score corresponding to each data source includes: Obtaining a weight score of the data source based on the product of the influence weight corresponding to the data source and the theoretical score corresponding to the data source; Based on the sum of the weighted scores corresponding to each of the data sources, a theoretical score of the delivered vehicle model is obtained.

6. The word-of-mouth data analysis method according to claim 1, characterized in that: The method further comprises: A word-of-mouth analysis result is generated based on the word-of-mouth score of each of the delivered models.

7. A word-of-mouth data analysis device, characterized in that: include: An acquisition unit, used to acquire public opinion data to be analyzed and vehicle delivery data; wherein the public opinion data to be analyzed does not include vehicle model information, and the vehicle delivery data includes delivery information of at least one delivered vehicle model; A processing unit, configured to perform data classification on the public opinion data to be analyzed acquired by the acquisition unit, and obtain an actual classification result of the public opinion data to be analyzed; The processing unit is also used to generate a reputation score for each delivered vehicle model based on historical public opinion data, the vehicle delivery data and the actual classification result; wherein the historical public opinion data includes public opinion data that does not mention the delivered vehicle model and the actual classification result is the target classification result.

8. An electronic device, characterized in that: include: A memory and a processor, wherein the memory is used to store a computer program; and the processor is used to enable the electronic device to implement the word-of-mouth data analysis method described in any one of claims 1 to 7 when executing the computer program.

9. A computer-readable storage medium, characterized in that: include: The computer-readable storage medium stores a computer program, and when the computer program is executed by a computing device, the computing device implements the word-of-mouth data analysis method according to any one of claims 1 to 7.

10. A vehicle, comprising the word-of-mouth data analysis device according to claim 8.

Images