Evaluation statement processing method and device, electronic equipment and storage medium

Abstract

The invention provides an evaluation statement processing method. The method relates to the technical field of natural language processing. The method comprises the steps of obtaining keywords in at least two evaluation statements; determining evaluation types of keywords in the at least two evaluation statements and detail degree information of the keywords; determining value values of the at least two evaluation statements based on the evaluation types and the detail degree information of keywords in the evaluation statements; and selecting an evaluation statement from the at least two evaluation statements according to the value numerical value, and taking at least one selected evaluation statement as a target evaluation statement for display. According to the embodiment of the invention, the evaluation with more reference value for the user can be screened out from a plurality of evaluations, and the evaluations are displayed to the user, so that the evaluation watching experience of the user can be improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of natural language processing (NLP), and in particular to a method, device, electronic device and storage medium for processing evaluation sentences. Background Art

[0002] With the development of Internet technology, more and more users are evaluating the objects they experience or purchase on the web. The evaluations of different users may vary greatly, so there may be a variety of evaluations on the web.

[0003] When there are a lot of reviews, there will be a large number of user-oriented reviews. In this way, there will be a lot of invalid information for users who read the reviews, and it is difficult for users to find useful reviews. In addition, the review information mixed with various emotional factors is large and complex, resulting in the review information providing little reference value to users and poor user experience. Summary of the invention

[0004] In view of this, the purpose of the present disclosure is to propose a method, device, electronic device and storage medium for processing evaluation statements, which can solve the existing problems in a targeted manner.

[0005] Based on the above-mentioned purpose, in a first aspect, the present disclosure proposes a method for processing evaluation statements, including: obtaining keywords in at least two evaluation statements; determining the evaluation type of the keywords in the at least two evaluation statements, and the detail level information of the keywords; determining the value values ​​of the at least two evaluation statements based on the evaluation type and the detail level information of the keywords in the evaluation statements; and selecting an evaluation statement from the at least two evaluation statements according to the value values, and using at least one selected evaluation statement as a target evaluation statement for display.

[0006] In a second aspect, a device for processing evaluation statements is also provided, comprising: a keyword acquisition unit, configured to acquire keywords in at least two evaluation statements; a type determination unit, configured to determine the evaluation type of the keywords in the at least two evaluation statements, and detail level information of the keywords; a value determination unit, configured to determine the value values ​​of the at least two evaluation statements based on the evaluation type and detail level information of the keywords in the evaluation statements; and a visualization unit, configured to select evaluation statements from the at least two evaluation statements according to the value values, and use at least one selected evaluation statement as a target evaluation statement for display.

[0007] In a third aspect, an electronic device is also provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in the first aspect.

[0008] According to a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, and the program is executed by a processor to implement any method described in the first aspect.

[0009] In general, the present disclosure has at least the following beneficial effects: it can screen out evaluations that are more valuable to users from a large number of evaluations, and display these evaluations to users, which helps to improve the user's experience of viewing evaluations. In addition, the present application can distinguish each evaluation statement by value value and display it to users in a targeted manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the multiple drawings represent the same or similar parts or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings only depict some embodiments disclosed in accordance with the present disclosure and should not be regarded as limiting the scope of the present disclosure.

[0011] Figure 1 A flow chart showing a method for processing evaluation statements of the present disclosure is shown;

[0012] Figure 2 A flowchart of determining at least one selected evaluation statement in the processing of evaluation statements according to an embodiment of the present disclosure is shown;

[0013] Figure 3 Another flow chart showing the method for processing evaluation statements of the present disclosure;

[0014] Figure 4 A schematic diagram showing a processing device for an evaluation statement of the present disclosure is shown;

[0015] Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present disclosure is shown;

[0016] Figure 6 A schematic diagram of a storage medium provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0017] The present disclosure is further described in detail below in conjunction with the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0018] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present disclosure may be combined with each other. The present disclosure will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0019] Figure 1 A method for processing an evaluation statement of the present disclosure is shown. In an embodiment of the present disclosure, the method comprises the following steps:

[0020] Step S101, obtaining keywords in at least two evaluation sentences.

[0021] In this embodiment, the execution subject of the processing method for executing the evaluation statement can determine the keywords in at least two evaluation statements. The keywords in this embodiment are not only meaningful words in the evaluation statement, but also can represent emotional tendencies. Emotional tendencies can include positive, negative and neutral. For example, keywords with positive emotional tendencies can include "the effect is too good." Keywords with negative emotional tendencies can include "the service attitude is very poor." Keywords with neutral emotional tendencies can include "the effect is okay."

[0022] Step S102: determining the evaluation type of the keywords in the at least two evaluation sentences and the detail level information of the keywords.

[0023] In this embodiment, the execution subject can determine the evaluation type of the keyword in the evaluation sentence. Here, the evaluation sentence for which the evaluation type and detail level information are to be determined is each evaluation sentence of the at least two evaluation sentences to be processed.

[0024] In practice, each evaluation type is used to characterize the user's experience of one aspect of the evaluated object. Each evaluation type corresponds to one evaluation aspect. For example, when the evaluated object is an experienced project (such as a medical project, medical beauty project, etc.), the evaluation type may include at least one of the following: project effect, doctor's skills, service attitude and hospital environment.

[0025] The evaluation statement is a description of the experience of the evaluated object. For example, the evaluation statement may be "I have become younger", "The doctor has good skills", etc.

[0026] The detail level information is used to characterize the specific level described by the keyword. The above-mentioned execution subject can determine the detail level information of the keyword. Specifically, the keyword can describe details such as "pores become smaller", or it can be a general summary such as "the effect is good". The keywords described in these two ways have different specific levels, that is, the detail level information of the keywords is different.

[0027] Step S103: determining the value values ​​of the at least two evaluation statements based on the evaluation type and the detail level information of the keywords in the evaluation statements.

[0028] In this embodiment, the execution subject may determine the value of each evaluation statement based on the evaluation type of a keyword in the evaluation statement and the detail level information of the keyword.

[0029] The value value is used to characterize the reference value of the evaluation statement for the user. In practice, the above-mentioned execution subject can adopt various methods to determine the value value of the evaluation statement based on the evaluation type of the keywords in the evaluation statement and the level of detail information of the keywords. For example, the above-mentioned execution subject can input the evaluation type and level of detail information of each keyword in the evaluation statement into a pre-trained model (such as a deep neural network) to obtain the value value of the evaluation statement. The model can use the evaluation type and level of detail information of the keywords in the evaluation statement to predict the value value of the evaluation statement.

[0030] Step S104: selecting an evaluation statement from the at least two evaluation statements according to the value value, and using at least one selected evaluation statement as a target evaluation statement for display.

[0031] In this embodiment, the execution subject may select an evaluation statement from the at least two evaluation statements according to the value value in various ways. For example, the execution subject may select an evaluation statement whose value value is greater than a preset value threshold from the at least two evaluation statements. After the selection, the execution subject may use at least one of the selected evaluation statements as a target evaluation statement for display.

[0032] The target evaluation statement is used to be displayed to the user. The target evaluation statement is displayed on the evaluation page loaded by the client, and the step of displaying the target evaluation statement may include: displaying the target evaluation statement in the evaluation page corresponding to the evaluated object of each target evaluation statement.

[0033] This embodiment can filter out evaluations that are more valuable to users from a large number of evaluations, and display these evaluations to users, which helps to improve the user's experience of viewing evaluations. In addition, this application can distinguish each evaluation statement by value value and display it to users in a targeted manner.

[0034] In some optional implementations of any embodiment of the present disclosure, the detail level information is a detail level value; the above-mentioned determination of the value values ​​of the at least two evaluation statements based on the evaluation type and the detail level information of the keywords in the evaluation statement may include: taking the priority corresponding to the evaluation type of the keywords in each evaluation statement as a weight, weighting the detail level values ​​of each keyword in the evaluation statement, and obtaining the value value of the evaluation statement.

[0035] In these implementations, the execution subject can determine the value of the evaluation statement by weighting. Specifically, the execution subject can use the priority corresponding to the evaluation type of the keyword as the weight to weight the detail value of each keyword in each evaluation statement. Each evaluation statement here refers to an evaluation statement in the at least two evaluation statements mentioned above. The weight represents the priority between the evaluation types, and the priority can reflect the importance of each evaluation type.

[0036] The larger the value of the level of detail, the greater the specificity of the keyword representation corresponding to the level of detail. For example, a keyword can include two levels of detail: detailed description and rough description. The level of detail value corresponding to the detailed description is 10, and the level of detail value corresponding to the rough description is 5. The priorities of the four evaluation types, project effect, doctor's skills, service attitude, and hospital environment, are arranged from high to low, and the corresponding weights are 4, 3, 2, and 1, respectively. If there are two keywords in an evaluation sentence, the first keyword is "pores are shrinking", which is a detailed description, so the corresponding level of detail value is 10. And the evaluation type of this keyword is project effect, the priority is 4, that is, the weight is 4. The second keyword is "good service attitude", which is a rough description, so the corresponding level of detail value is 5. And the evaluation type of this keyword is service attitude, the priority is 2, that is, the weight is 2. Weighting this evaluation sentence is:

[0037] 10×4+5×2=50.

[0038] Therefore, the value of this evaluation statement is 50.

[0039] These implementations can take into account the importance of different evaluation types and accurately determine the value of different evaluation statements by weighting the detail levels.

[0040] Figure 2The method for processing evaluation sentences in an embodiment of the present disclosure shows the steps of taking at least one selected evaluation sentence as a target evaluation sentence for display: S201, calling a pre-trained sentiment analysis model to determine the sentiment tendency of the selected evaluation sentence; S202, determining, among the selected evaluation sentences, evaluation sentences with positive sentiment tendency as candidate evaluation sentences, and determining the evaluation sentence for display among the candidate evaluation sentences.

[0041] In this embodiment, the sentiment tendency of the target evaluation sentence for display is positive. The above-mentioned execution subject can call the sentiment analysis model to analyze the sentiment tendency of the evaluation sentence as a whole and filter out the evaluation sentence with positive sentiment tendency. Specifically, the sentiment analysis model is a natural language processing model that can determine the sentiment tendency of the evaluation sentence, such as SnowNLP.

[0042] In practice, the execution subject may determine the evaluation statements for display from the candidate evaluation statements in various ways, such as directly determining all the candidate evaluation statements as the evaluation statements for display.

[0043] This embodiment uses a sentiment analysis model to accurately determine evaluation sentences with positive sentiment tendencies.

[0044] In some optional application scenarios of these implementations, the training steps adopted by the sentiment analysis model may include: extracting evaluation sentences as positive samples with positive sentiment tendencies from evaluation texts with user scores greater than or equal to a first score value, and extracting evaluation sentences as negative samples with negative sentiment tendencies from evaluation texts with user scores less than or equal to a second score value, wherein the first score value is greater than the second score value; using the positive samples and the negative samples to train the sentiment analysis model to be trained to obtain the trained sentiment analysis model.

[0045] In these application scenarios, there are corresponding user ratings for the evaluation sentences. The above-mentioned execution subject or other electronic device can use the evaluation texts with high user ratings as positive samples, and the evaluation texts with low user ratings as negative samples to train the sentiment analysis model to be trained. For example, the first rating value can be 4.0, and the second rating value can be 1.0. Evaluation texts greater than or equal to 4.0 can be used as positive samples, and evaluation texts less than or equal to 1.0 can be used as negative samples.

[0046] In practice, positive and negative samples can be saved in different files, and each sample can correspond to a line in the file.

[0047] These application scenarios can use user evaluation texts to train sentiment analysis models, so that the sentiment analysis models can learn various features of the evaluation texts, which is conducive to improving the accuracy of the sentiment analysis models in analyzing evaluation sentences.

[0048] The present disclosure also provides another embodiment to implement the above-mentioned use of at least one selected evaluation sentence as a target evaluation sentence for display: for each selected evaluation sentence, the sentiment tendency of the keywords in the evaluation sentence is searched in the sentiment tendency dictionary, and the sentiment tendencies of the keywords in the evaluation sentence are summarized to obtain the sentiment tendency of the evaluation sentence; among the selected evaluation sentences, the evaluation sentences with positive sentiment tendencies are determined as candidate evaluation sentences, and the evaluation sentences for display are determined among the candidate evaluation sentences.

[0049] In this embodiment, as another way to determine the sentiment tendency, the above-mentioned execution subject can also determine the sentiment tendency of each keyword in the evaluation sentence through the sentiment tendency dictionary. The sentiment tendency dictionary can have a correspondence between keywords and sentiment tendencies. In this way, the above-mentioned execution subject can find the sentiment tendencies corresponding to each keyword in the evaluation sentence in the sentiment tendency dictionary. Afterwards, the above-mentioned execution subject can summarize the sentiment tendencies of each keyword.

[0050] For example, the evaluation sentence is "The effect is not obvious", which may include the keywords "effect", "not", and "obvious". The sentiment tendencies of these three keywords are: the sentiment tendency of "effect" is positive, the sentiment tendency of "not" is negative, and the sentiment tendency of "obvious" is positive. After aggregation, the summary result of the evaluation sentence is positive, so the sentiment tendency of the evaluation sentence is positive.

[0051] This embodiment can provide another way to analyze sentiment tendency besides the sentiment analysis model, and accurately find the sentiment tendency corresponding to the evaluation sentence through the sentiment dictionary.

[0052] In some optional implementations of any of the above embodiments, the two methods for determining the sentiment tendency provided above can also be used in combination. For example, the sentiment analysis model can be first applied to the sentiment tendency dictionary to determine the sentiment tendency of each keyword in the evaluation sentence, thereby determining the sentiment tendency of the evaluation sentence and obtaining a first evaluation sentence with a positive sentiment tendency; then the sentiment analysis model is used to analyze the sentiment tendency of the first evaluation sentence, and a second evaluation sentence is obtained based on the sentiment analysis result; or, the sentiment analysis model can be first used to analyze to obtain a first result, and then the sentiment tendency dictionary can be used to analyze the first result; or, the sentiment analysis model and the sentiment tendency dictionary can be used to analyze the evaluation sentence respectively, and then the evaluation sentence with both analysis results being positive is selected as the final result.

[0053] In some optional implementations of any of the above embodiments, determining an evaluation statement for display from candidate evaluation statements includes: determining an evaluation statement for display from candidate evaluation statements whose number of characters is within a preset character number interval, wherein the lower limit of the preset character number interval is a first preset number of characters, and the upper limit of the preset character number interval is a second preset number of characters.

[0054] In these implementations, the execution subject can select evaluation sentences with a moderate length from the candidate evaluation sentences. For example, the lower limit of the preset character number interval is also the first preset character number, which can be 2. The upper limit of the preset character number interval is also the second preset character number, which can be 20.

[0055] It should be noted that the first preset number of characters and the second preset number of characters may be default lengths or lengths set by the user.

[0056] These implementation methods can find reviews with moderate sentence lengths based on the number of characters and remove reviews that are too long or too short, thereby improving the aesthetics of the review information displayed on the page.

[0057] In some optional implementations of any of the above embodiments, before determining at least one evaluation type for each of at least two evaluation statements to be processed, the method further includes: extracting the at least two evaluation statements from one or more evaluation texts to be processed, wherein each of the one or more evaluation texts to be processed includes at least one evaluation statement.

[0058] In these implementations, the above-mentioned execution entity can extract the at least two evaluation sentences from each evaluation text to be processed in various ways, for example, inputting each evaluation text to be processed into a preset sentence extraction model (such as a deep neural network) and obtaining at least two evaluation sentences output from the model.

[0059] These implementations can first extract the evaluation sentences from the evaluation text before processing the evaluation sentences, thereby facilitating the smooth screening and display of valuable evaluation sentences.

[0060] In some optional application scenarios of the above embodiments, the step of determining the one or more evaluation texts to be processed includes: extracting evaluation texts whose user scores reach a third score value from an evaluation text library as the one or more evaluation texts to be processed.

[0061] In these optional application scenarios, the execution subject can extract the evaluation texts with high user ratings. In practice, if the highest user rating is 5 points, 4.5 can be used as the third rating value to extract the evaluation texts with user ratings greater than 4.5.

[0062] These optional application scenarios can improve the positive attributes of the evaluation text and help users find high-quality experience projects.

[0063] In some optional application scenarios of any of the above embodiments, selecting an evaluation statement from the at least two evaluation statements based on the value value includes: filtering out the evaluation statements from the at least two evaluation statements in order from large to small of the value values ​​of the evaluation statements, wherein the number of the filtered evaluation statements is less than or equal to a preset number threshold.

[0064] In these optional application scenarios, the above-mentioned execution entity can extract the at least two evaluation sentences from the evaluation text to be processed in various ways. For example, each evaluation text among the multiple evaluation texts to be processed, as well as the value of the evaluation text, is input into a preset sentence extraction model (such as a deep neural network), and at least two evaluation sentences are output from the model.

[0065] The execution subject can sort the evaluation statements according to the value values ​​from large to small or from small to large to obtain the sorting results. And use the sorting results to filter out the evaluation statements in order from large to small. Specifically, the evaluation statements with value values ​​greater than the value threshold among these evaluation statements can be used as the evaluation statements obtained by screening. Alternatively, in order from large to small, a preset number of evaluation statements with the largest value values ​​among these evaluation statements can be used as the results obtained by screening. The number of screened evaluation statements can be limited, such as less than or equal to 3.

[0066] These application scenarios can accurately filter out evaluation sentences with high value values ​​by arranging them in order of value values.

[0067] Optionally, the evaluation statements used for screening may be derived from one evaluation text or from multiple evaluation texts. If the evaluation statements used for screening are derived from one evaluation text, at least one evaluation statement in the evaluation text is sorted from large to small according to the value values ​​of the evaluation statements, thereby screening out high-quality evaluation statements in the evaluation text. These optional application scenarios make full use of each evaluation text by determining the evaluation statements in each evaluation text.

[0068] For the evaluation sentences used for screening that come from multiple evaluation texts, these optional application scenarios may not be limited to a single evaluation text, but may focus on all the evaluation texts and perform value screening on each evaluation sentence in these evaluation texts, thereby helping to find evaluation sentences with higher values.

[0069] In some optional implementations of any of the above embodiments, the determining of the keywords in the at least two evaluation sentences may include: searching for keywords in the at least two evaluation sentences that exist in a forward keyword library.

[0070] In these optional implementations, the keywords can be used to characterize positive sentiment tendencies. In the preset positive keyword library, there may be keywords with positive sentiment tendencies. The above-mentioned execution subject may match each word in at least two evaluation sentences with the keywords in the positive keyword library, so as to find the same words as the keywords in the positive keyword library. Afterwards, the above-mentioned execution subject may use the found words as keywords in at least two evaluation sentences.

[0071] These implementations can comprehensively and accurately find keywords with positive sentiment through a preset positive keyword library. In addition, the keywords in the present disclosure represent positive sentiment, which allows users to see which experience projects have more positive reviews, which is conducive to customers choosing high-quality experience projects.

[0072] Optionally, extracting the at least two evaluation sentences from the multiple evaluation texts to be processed may include: for each evaluation text in the multiple evaluation texts to be processed, dividing the evaluation text into multiple sentences according to punctuation marks; removing stop words in the multiple sentences to obtain the at least two evaluation sentences.

[0073] The above execution entity can segment the evaluation text into sentences and remove stop words, thereby removing invalid information and helping to quickly and accurately find keywords in the sentence.

[0074] like Figure 3 As shown, the present disclosure also provides another flow chart of a method for processing an evaluation statement.

[0075] The embodiment of the present application provides a processing device for evaluation statements, which is used to execute the processing method for evaluation statements described in the above embodiment, such as Figure 4As shown, the device includes: a keyword acquisition unit, configured to acquire keywords in at least two evaluation sentences; a type determination unit, configured to determine the evaluation type of the keywords in the at least two evaluation sentences, and the detail level information of the keywords; a value determination unit, configured to determine the value values ​​of the at least two evaluation sentences based on the evaluation type and the detail level information of the keywords in the evaluation sentences; and a visualization unit, configured to select evaluation sentences from the at least two evaluation sentences according to the value values, and use at least one selected evaluation sentence as a target evaluation sentence for display.

[0076] Optionally, the detail level information is a detail level value; the value determination unit is configured to execute the evaluation type based on the keywords in the evaluation statement and the detail level information in the following manner to determine the value values ​​of the at least two evaluation statements: taking the priority corresponding to the evaluation type of the keyword in each evaluation statement as a weight, weighting the detail level values ​​of each keyword in the evaluation statement to obtain the value value of the evaluation statement, wherein the weight represents the priority between the evaluation types.

[0077] Optionally, the visualization unit is further configured to execute the method of using at least one selected evaluation sentence as a target evaluation sentence for display in the following manner: calling a pre-trained sentiment analysis model to determine the sentiment tendency of the selected evaluation sentence; among the selected evaluation sentences, determining the evaluation sentences with positive sentiment tendency as candidate evaluation sentences, and determining the evaluation sentences for display among the candidate evaluation sentences.

[0078] Optionally, the training steps adopted by the sentiment analysis model include: extracting evaluation sentences as positive samples with positive sentiment tendency from evaluation texts with user scores greater than or equal to a first score value, and extracting evaluation sentences as negative samples with negative sentiment tendency from evaluation texts with user scores less than or equal to a second score value, wherein the first score value is greater than the second score value; using the positive samples and the negative samples to train the sentiment analysis model to be trained to obtain the trained sentiment analysis model.

[0079] Optionally, determining an evaluation statement for display from candidate evaluation statements includes: determining a candidate evaluation statement whose number of characters is within a preset character number range as the at least one selected evaluation statement, wherein the lower limit of the preset character number range is a first preset number of characters, and the upper limit of the preset character number range is a second preset number of characters.

[0080] Optionally, the visualization unit is further configured to execute the method of taking at least one selected evaluation sentence as the target evaluation sentence for display in the following manner: for each selected evaluation sentence, search the sentiment tendency of the keywords in the evaluation sentence in the sentiment tendency dictionary, summarize the sentiment tendency of each keyword in the evaluation sentence, and obtain the sentiment tendency of the evaluation sentence; among the selected evaluation sentences, determine the evaluation sentences with positive sentiment tendency as candidate evaluation sentences, and determine the evaluation sentences for display among the candidate evaluation sentences.

[0081] Optionally, before obtaining keywords in at least two evaluation sentences, the above-mentioned device also includes: an extraction unit, configured to extract the at least two evaluation sentences from one or more evaluation texts to be processed, wherein each evaluation text in the one or more evaluation texts to be processed includes at least one evaluation sentence.

[0082] Optionally, the target determination unit is further configured to execute the selection of an evaluation statement from the at least two evaluation statements based on the value value in the following manner: filtering evaluation statements from the evaluation statements in the evaluation text in descending order of the value values ​​of the evaluation statements, wherein the number of the filtered evaluation statements is less than or equal to a preset number threshold.

[0083] Optionally, the target determination unit is further configured to perform the filtering of evaluation statements from the evaluation statements in the evaluation text in the order of the value values ​​of the evaluation statements from large to small in the following manner: for at least one evaluation statement contained in each evaluation text, the filtering of evaluation statements from the evaluation text in the order of the value values ​​of the evaluation statements from large to small.

[0084] Optionally, the target determination unit is further configured to perform the filtering of evaluation statements from the evaluation statements in the evaluation texts in the order of the value values ​​of the evaluation statements from large to small in the following manner: for at least one evaluation statement included in the multiple evaluation texts, filtering the evaluation statements from the evaluation statements in the multiple evaluation texts in the order of the value values ​​of the evaluation statements from large to small.

[0085] Optionally, the step of determining the one or more evaluation texts to be processed includes: extracting evaluation texts whose user scores reach a third score value from an evaluation text library as the one or more evaluation texts to be processed.

[0086] Optionally, the keyword acquisition unit is further configured to perform the acquisition of keywords in the at least two evaluation sentences in the following manner: searching for keywords existing in a forward keyword library in the at least two evaluation sentences.

[0087] The method for processing evaluation statements provided in the above-mentioned embodiments of the present disclosure is based on the same inventive concept as the method for processing evaluation statements provided in the embodiments of the present disclosure, and has the same beneficial effects as the method adopted, run or implemented by the application program storing the same.

[0088] The embodiment of the present disclosure also provides an electronic device corresponding to the method for processing the evaluation statement provided in the above embodiment, so as to execute the above method for processing the evaluation statement. The embodiment of the present disclosure is not limited.

[0089] Please refer to Figure 5 , which shows a schematic diagram of an electronic device provided by some embodiments of the present disclosure. Figure 5 As shown, the electronic device 50 includes: a processor 500, a memory 501, a bus 502 and a communication interface 503, and the processor 500, the communication interface 503 and the memory 501 are connected via the bus 502; the memory 501 stores a computer program that can be run on the processor 500, and the processor 500 executes the method provided in any of the aforementioned embodiments of the present disclosure when running the computer program.

[0090] The memory 501 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 503 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0091] The bus 502 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The memory 501 is used to store programs, and the processor 500 executes the programs after receiving execution instructions. The processing method of the evaluation statement disclosed in any implementation of the aforementioned embodiment of the present disclosure may be applied to the processor 500, or implemented by the processor 500.

[0092] The processor 500 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit in the processor 500 or the instruction in the form of software. The above processor 500 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present disclosure can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in conjunction with the embodiments of the present disclosure can be directly embodied as a hardware decoding processor to be executed, or the hardware and software modules in the decoding processor can be executed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 501, and the processor 500 reads the information in the memory 501 and completes the steps of the above method in combination with its hardware.

[0093] The electronic device provided by the embodiment of the present disclosure and the method for processing the evaluation statement provided by the embodiment of the present disclosure are based on the same inventive concept and have the same beneficial effects as the method adopted, run or implemented by them.

[0094] The present disclosure also provides a computer-readable storage medium corresponding to the method for processing the evaluation statement provided in the above embodiment. Figure 6 The computer-readable storage medium shown is a CD 60 on which a computer program (ie, a program product) is stored. When the computer program is executed by a processor, the processing method of the evaluation statement provided in any of the aforementioned embodiments will be executed.

[0095] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical or magnetic storage media, which are not listed here one by one.

[0096] The computer-readable storage medium provided by the above-mentioned embodiment of the present disclosure and the method for processing evaluation statements provided by the embodiment of the present disclosure are based on the same inventive concept, and have the same beneficial effects as the method adopted, run or implemented by the application program stored therein.

[0097] It should be noted that:

[0098] In the above sentences, the terms "comprises", "includes" 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, an element defined by "including one..." does not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be noted that the scope of the methods and devices in the embodiments of the present disclosure is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0099] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present disclosure, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present disclosure.

[0100] The embodiments of the present disclosure are described above in conjunction with the accompanying drawings, which are only specific implementation modes of the present disclosure. However, the present disclosure is not limited to the above-mentioned specific implementation modes. The above-mentioned specific implementation modes are merely illustrative and not restrictive. Under the guidance of the present disclosure, ordinary technicians in this field can also make many forms without departing from the purpose of the present disclosure and the scope of protection of the claims, all of which are within the protection of the present disclosure.

Claims

1. A method for processing an evaluation statement, characterized in that: include: Obtain keywords from at least two evaluation sentences; Determining evaluation types of keywords in the at least two evaluation sentences and detail level information of the keywords; Determining the value values ​​of the at least two evaluation statements based on the evaluation type and the detail level information of the keywords in the evaluation statements; An evaluation statement is selected from the at least two evaluation statements according to the value value, and at least one of the selected evaluation statements is used as a target evaluation statement for display.

2. The processing method according to claim 1, characterized in that: The detail level information includes a detail level value; The step of determining the value values ​​of the at least two evaluation statements based on the evaluation type and the detail level information of the keywords in the evaluation statements includes: The priority corresponding to the evaluation type of the keyword in each evaluation sentence is used as the weight, and the detail degree value of each keyword in the evaluation sentence is weighted to obtain the value value of the evaluation sentence.

3. The processing method according to claim 1, characterized in that: The step of using at least one selected evaluation sentence as a target evaluation sentence for display includes: Call the pre-trained sentiment analysis model to determine the sentiment tendency of the selected evaluation sentence; Among the selected evaluation sentences, evaluation sentences with positive sentiment tendencies are determined as candidate evaluation sentences, and target evaluation sentences for display are determined among the candidate evaluation sentences.

4. The processing method according to claim 3, characterized in that: The training steps adopted by the sentiment analysis model include: Extracting evaluation sentences as positive samples with positive sentiment tendency from evaluation texts with user scores greater than or equal to a first score value, and extracting evaluation sentences as negative samples with negative sentiment tendency from evaluation texts with user scores less than or equal to a second score value, wherein the first score value is greater than the second score value; The positive samples and the negative samples are used to train the sentiment analysis model to be trained to obtain a trained sentiment analysis model.

5. The processing method according to claim 1, characterized in that: The step of using at least one selected evaluation sentence as a target evaluation sentence for display includes: For each selected evaluation sentence, the sentiment tendency of the key words in the evaluation sentence is searched in the sentiment tendency dictionary, and the sentiment tendency of each key word in the evaluation sentence is summarized to obtain the sentiment tendency of the evaluation sentence; Among the selected evaluation sentences, evaluation sentences with positive sentiment tendencies are determined as candidate evaluation sentences, and target evaluation sentences for display are determined among the candidate evaluation sentences.

6. The processing method according to any one of claims 3 to 5, characterized in that: Determine the target evaluation sentence for display from the candidate evaluation sentences, including: A target evaluation sentence for display is determined among candidate evaluation sentences whose number of characters is within a preset character number interval, wherein the lower limit of the preset character number interval is a first preset number of characters and the upper limit of the preset character number interval is a second preset number of characters.

7. The processing method according to any one of claims 1 to 5, characterized in that: Before obtaining the keywords in at least two evaluation sentences, the method further includes: The at least two evaluation sentences are extracted from one or more evaluation texts to be processed, wherein each evaluation text in the multiple evaluation texts to be processed includes at least one evaluation sentence.

8. The processing method according to claim 7, characterized in that: The step of selecting an evaluation statement from the at least two evaluation statements according to the value value includes: Evaluation statements are screened from the at least two evaluation statements in descending order of the value values ​​of the evaluation statements, wherein the number of the screened evaluation statements is less than or equal to a preset number threshold.

9. The processing method according to claim 7, characterized in that: The step of determining one or more evaluation texts to be processed comprises: From the evaluation text library, evaluation texts with user ratings reaching a third rating value are extracted as the one or more evaluation texts to be processed.

10. The processing method according to any one of claims 1 to 5, characterized in that: The obtaining of keywords in at least two evaluation sentences includes: In the at least two evaluation sentences, keywords existing in the forward keyword library are searched.

11. A device for processing evaluation statements, characterized in that: include: A keyword acquisition unit, configured to acquire keywords from at least two evaluation sentences; a type determination unit configured to determine the evaluation type of the keyword in the at least two evaluation sentences, and detail level information of the keyword; a value determination unit configured to determine value values ​​of the at least two evaluation sentences based on the evaluation type and the detail level information of the keywords in the evaluation sentences; The visualization unit is configured to select an evaluation statement from the at least two evaluation statements according to the value value, and use at least one selected evaluation statement as a target evaluation statement for display.

12. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: The processor runs the computer program to implement the method according to any one of claims 1-10.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the method according to any one of claims 1 to 10.

Images