WO2020057237A1 - Influence detection method applicable to object of interest, and electronic terminal and storage medium - Google Patents

Abstract

An influence detection method of an object of interest, and an electronic terminal and a storage medium. The method comprises: performing search by using a search word to obtain a set of a search result item; and calculating an influence evaluation parameter of the object of interest obtained on the basis of the search result item associated with the object of interest in the set, the influence evaluation parameter comprising a word-of-mouth degree and/or visibility; wherein the word-of-mouth degree represents a degree of favorable comment of the object of interest obtained on the basis of the set; and the visibility represents the degree of ranking of the search result item associated with the object of interest in the set. The method analyzes the influence intensity of the object of interest on the basis of a plurality of influence evaluation parameters such as word-of-mouth degree, visibility, popular degree value, and concentration value, thereby effectively and comprehensively evaluating the network influence of a brand.

Description

Impact detection method, electronic terminal and storage medium suitable for attention object Technical field

The present invention relates to the field of network technology, and in particular, to an influence detection method, an electronic terminal, and a storage medium suitable for an object of interest.

Background technique

Nowadays, people's lives are more and more inseparable from the Internet. Consumers are immersed in the world of the Internet. Internet advertising is also full of almost all network platforms, such as videos, portals, vertical portals, search, and so on. What is easy to overlook is that content is more meaningful than advertisements, because consumers are not willing to look at advertisements, but rather want to find the content they want to know from other people's evaluations, views, opinions, text or science.

However, the existing technology cannot provide an effective solution to enable consumers to understand the factors affecting a brand's influence such as brand awareness and word of mouth when they access Internet content.

Summary of the Invention

In view of the shortcomings of the prior art described above, an object of the present invention is to provide an influence detection method, an electronic terminal, and a storage medium suitable for an object of interest, which are used to solve the existing technology and cannot provide an effective solution to enable consumption. Understand the technical aspects of brand influence when they access Internet content.

In order to achieve the above object and other related objects, the present invention provides an influence detection method suitable for an object of interest, which includes obtaining a set of search result entries by using a search term; and calculating a search based on the set that is associated with the object of interest The influence evaluation parameters of the object of interest obtained from the result entry include word-of-mouth and / or visibility; wherein the word-of-mouth expresses the praise degree of the object of interest obtained based on the set; the visibility expresses Describes the extent to which the search results associated with the object of interest rank in the set.

In an embodiment of the present invention, the object of interest includes any one or more of a brand name, a product name, a company CEO name, a company slogan, or a competing product; the search terms include: consumer demand words , Brand words, brand extension words, corporate words, competitive words, any one or more combinations.

In an embodiment of the present invention, the method of calculating the word-of-mouth reputation includes calculating the word-of-mouth reputation of the attention object obtained based on the direct display content of search result entries associated with the attention object in the set, and / Or calculate the reputation of the object of interest obtained based on the indirect display of content of the search result entry associated with the object of interest in the collection; wherein the directly displayed content includes the title and summary of the search result entry, The indirect display content includes the content of the manuscript of the search result entry.

In an embodiment of the present invention, the calculating the word-of-mouth degree of the object of interest obtained based on the direct display content of search result entries associated with the object of interest in the collection, and the calculation method specifically includes: Describe the sentiment analysis of the subject of interest in the context of the title and summary of each search result entry, and judge the word of mouth of the subject of interest based on each search result entry based on the sentiment analysis result; based on the sentiment analysis result of each search result entry Closeness to positive emotions, screen out search result items that meet the word-of-mouth requirements; calculate the ratio of the number of search result items that meet the word-of-mouth requirements to the total number of search result entries associated with the object of interest, and use the calculated The scale determines the word of mouth of the object of interest.

In an embodiment of the present invention, the calculating the word-of-mouth reputation of the concerned object obtained based on the indirect display of search result entries associated with the concerned object in the collection, and the calculation method thereof specifically includes: determining the Describe the positions of the subject of interest in the manuscript content of the search result entry; perform an sentiment analysis on the context of the subject of interest in the respective positions and obtain the sentiment analysis results of the respective positions, and correspond to the positions Comprehensive analysis of the sentiment analysis results to get the sentiment analysis results of the search result item; according to the closeness of the sentiment analysis results of each search result item to the positive emotions, assign a corresponding weight to each search result item; select the weight value that meets word of mouth requirements And calculate the ratio of the number of search result entries to the total number of search result entries associated with the object of interest, and the calculated proportion is used to determine the reputation of the object of interest.

In an embodiment of the present invention, the sentiment analysis results of each location include a non-negative evaluation and a negative evaluation, and the non-negative evaluation includes a positive evaluation and a neutral evaluation; wherein: if the sentiment analysis results of all the locations are non-negative evaluations Negative evaluation, then determine that the sentiment analysis result of the search result item is good and give the search result item a high weight; if the sentiment analysis result is a non-negative evaluation, the number of positions is greater than the number of the negative analysis result. , Determine that the sentiment analysis result of the search result item is better and assign a higher weight to the search result item; if the number of positions where the sentiment analysis result is a non-negative evaluation is equal to the number of positions where the sentiment analysis result is a negative evaluation, Then determine that the sentiment analysis result of the search result item is medium and give the search result item a medium weight; if the number of positions where the sentiment analysis result is non-negative evaluation is less than the number of positions where the sentiment analysis result is negative evaluation, then determine the The sentiment analysis result of the search result entry is poor and given the search result bar A lower weight; if all the emotional position of the analysis results are negative comments, the search results is determined entry emotional results for the poor and give to the search result entry a low weight.

In an embodiment of the present invention, the reputation is:

In an embodiment of the present invention, a manner of analyzing the visibility of search result entries associated with the object of interest in the search result entry set includes:

Among them, pct _i represents the visibility of the i-th search result entry in the search result entry set, x _i represents the assignment of the i-th search result entry in the page, and n represents n search result entries; let a preset number of searches The result entries are classified into the same search page, and the weighted visibility of each search result entry is calculated according to the weight of the search page in all search pages.

among them,

weight _j represents the weight of the j-th search page among all search pages, and x _ij represents the assignment of the i-th search result entry in the j-th search page.

In an embodiment of the present invention, the influence evaluation parameter of the object of interest further includes a concentration, and the concentration is used to indicate a proportion of search result entries associated with the object of interest in the set. Calculations include:

Among them, y ₁ represents search result entries associated with the object of interest, Counts of y ₁ represents the number of search result entries associated with the object of interest; Total y represents the total number of search result entries in the set.

In an embodiment of the present invention, the method includes: calculating a word-of-mouth reputation obtained by a same website based on a plurality of different search terms based on a preset website; and obtaining a word of mouth based on a plurality of different search terms based on a preset website based on the target of interest Of word of mouth and the popularity of each search term on the website, calculate the word of mouth index of the object of interest based on the website's word-of-mouth index; calculate the word-of-mouth index of the website and the usage rate of the website within a preset time period to calculate the The object of interest is based on the word-of-mouth influence index of the website; the word-of-mouth influence index of each website on the network is added up to generate a corresponding word-of-mouth influence index; the word-of-mouth influence index is used to indicate that the object of interest is in Word of mouth performance on the entire network.

In an embodiment of the present invention, the method: the calculation formula of the website word of mouth index is:

Among them, pct _p1 , ..., pct _pn indicate that the object of interest is based on the word of mouth obtained by the website based on the 1st to nth search terms, and V1, ..., Vn indicate the popularity value of the 1st to n search terms on the website Wherein, the type of the popularity value includes any one or more combinations of search volume, attention volume, and popularity index; the calculation formula of the comprehensive influence index is: Web_Index _p = Web_pct _p * Web_Mount Percent; where , Web_Mount Percent indicates the usage rate of the website within a preset time period; the calculation formula of the comprehensive influence index of word of mouth is: Total_Web_Index _p = ΣWeb_Index _p .

In an embodiment of the present invention, the method includes: calculating the visibility obtained by the same interested object based on a plurality of different search terms based on a preset website; according to the concerned object based on a plurality of different search terms based on a preset website Visibility and popularity value of each search term on the website, calculate the object of interest based on the website visibility index of the website; according to the website visibility index and the usage rate of the website in a preset time period, the calculated The object of interest is based on the visibility impact index of the website; the visibility impact indices of each website on the network are summed to generate a corresponding visibility impact comprehensive index; the visibility impact comprehensive index is used for Represents the ranking performance of the object of interest on the entire network.

In an embodiment of the present invention, the calculation formula of the website visibility index is:

Among them, pct _s1 , ..., pct _sn indicate the visibility of the object of interest based on the website's first to nth search terms, and V1, ..., Vn indicate the popularity value of the first to n search terms on the website; Wherein, the type of the heat value includes any one or more combinations of search volume, attention volume, and heat index; the calculation formula of the visibility influence index is: Web_Index _s = Web_pct _s * Web_Mount Percent; wherein, Web_Mount Percent indicates the usage rate of the website within a preset period of time; the calculation formula of the visible influence comprehensive index is: Total_Web_Index _s = ΣWeb_Index _s .

In order to achieve the above-mentioned object and other related objects, the present invention provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method for detecting influences applicable to an object of interest is implemented.

To achieve the above object and other related objects, the present invention provides an electronic terminal including: a processor and a memory; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory so that the memory The terminal executes the influence detection method applicable to the object of interest.

As described above, the present invention provides an impact detection method, an electronic terminal, and a storage medium that are applicable to the object of interest, and has the following beneficial effects: The present invention analyzes the object of interest based on a number of impact evaluation parameters such as word of mouth, visibility, and heat value , Which effectively and comprehensively evaluates the brand ’s online influence.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic flowchart of an influence detection method applicable to an object of interest in an embodiment of the present invention.

FIG. 2 is a schematic flowchart of an influence detection method applicable to an object of interest in an embodiment of the present invention.

FIG. 3 is a schematic flowchart of an influence detection method applicable to an object of interest in an embodiment of the present invention.

FIG. 4 is a schematic flowchart of an influence detection method applicable to an object of interest in an embodiment of the present invention.

FIG. 5 is a schematic diagram showing the word-of-mouth degree of each search term on a Baidu website according to an embodiment of the present invention.

FIG. 6 is a two-dimensional matrix analysis diagram based on search volume and visibility according to an embodiment of the present invention.

FIG. 7 is a two-dimensional matrix analysis diagram based on search volume and visibility comparison values according to an embodiment of the present invention.

FIG. 8 is a two-dimensional matrix analysis diagram based on visibility and word of mouth in an embodiment of the present invention.

FIG. 9 is a multi-dimensional matrix analysis diagram based on search volume, visibility, and word of mouth in an embodiment of the present invention.

FIG. 10 is a schematic structural diagram of an electronic terminal according to an embodiment of the present invention.

detailed description

The following describes the embodiments of the present invention through specific specific examples. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through different specific implementations, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, in the case of no conflict, the following embodiments and features in the embodiments can be combined with each other.

It should be noted that, in the following description, reference is made to the accompanying drawings, which describe several embodiments of the present application. It should be understood that other embodiments may also be used, and changes in mechanical composition, structure, electrical, and operation may be made without departing from the spirit and scope of the application. The following detailed description should not be considered limiting, and the scope of the embodiments of this application is limited only by the claims of the published patent. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application. Space-related terms, such as "up", "down", "left", "right", "below", "below", "lower", "above", "upper", etc., can be used in the text to facilitate The relationship between one element or feature shown in the figure and another element or feature is explained.

Furthermore, as used herein, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context indicates otherwise. It should be further understood that the terms "comprising," " Includes "indicates the presence of stated features, operations, elements, components, items, categories, and / or groups, but does not exclude the presence of one or more other features, operations, elements, components, items, categories, and / or groups , Appearance or addition. The terms "or" and "and / or" as used herein are to be interpreted as inclusive or meaning any one or any combination. Therefore, "A, B or C" or "A, B and / Or C "means" any of the following: A; B; C; A and B; A and C; B and C; A, B, and C ". Only if the combination of elements, functions, or operations are inherent in some way An exception to this definition occurs only when the grounds are mutually exclusive.

The invention provides an impact detection method, an electronic terminal, and a storage medium suitable for an object of interest, which are based on a plurality of impact evaluation parameters such as word of mouth, visibility, heat value, and concentration value to analyze the impact intensity of the object of interest, thereby effectively The network influence of the brand has been comprehensively evaluated. The technical solution of the present invention will be explained below with reference to specific embodiments.

As shown in FIG. 1, a schematic flowchart of a method for detecting influence of a subject of interest in an embodiment of the present invention is shown. In this embodiment, the attention object refers to words that the user pays attention to and whose influence is to be analyzed, such as: brand name, product name, company CEO name, company slogan, slogan, or competing products. The influence detection method specifically performs the following steps:

S11: Use a search term to obtain a set of search result entries. The search terms include consumer demand words, brand words, brand extension words, business words, competitive words, and so on. Wherein, the consumer demand word refers to a search term generated based on consumer demand, such as "which cosmetic is more whitening" or "which new energy vehicle is more cost-effective", etc .; the brand term is based on a brand name The generated search terms, such as "L'Oreal Paris" or "Metersbonwe", etc .; the brand extension term refers to the search terms generated after expansion based on the brand name, such as "L'Oreal Paris is good" or "Apple Is your mobile phone expensive? "Etc .; the company term refers to a search term based on a company name or a product name or a company slogan associated with a brand; the competing term refers to a search term based on a competitor. It should be noted that the classification of search terms in this embodiment is for reference only and is not a limitation on the implementation method of the present invention.

Optionally, in this embodiment, the step of obtaining a search result entry by using a search term may be performed based on any website with a search function. The website with a search function is, for example, a search engine, a social media, an e-commerce platform, a blog or microblog platform, a news platform, a question and answer platform, a forum platform, or a video player platform. The search engine is, for example, Baidu website, Google website, Sogou website, etc .; the social media is WeChat platform or Momo platform, etc .; the e-commerce platform is, for example, JD.com, Tmall, Amazon, or Pinduoduo, etc. ; The blog or microblog platform is, for example, a Sina blog site, Sohu blog site, etc .; the news platform is, for example, today's headline website, Baidu news website or UC headline website, etc .; Wukong Q & A, etc .; the forum platform is, for example, Baidu Tieba or Douban, etc .; the video playback platform is, for example, iQiyi, Youku Video, Tencent Video, Youtube Video, etc. It should be noted that the website described in the present invention may be a portal website in a web page format or an APP website in the form of a client, which is not limited in the present invention.

S12: Calculate an influence evaluation parameter of the attention object obtained based on a search result entry associated with the attention object in the set, which includes a reputation and / or visibility. Wherein, the word-of-mouth reputation degree indicates the praise degree of the concerned object obtained based on the collection; and the visibility degree indicates the degree of ranking of the search result entries associated with the concerned object in the collection. It is worth noting that when analyzing the influence of an object of interest, it is necessary to understand not only whether the position of the object of interest is significant enough in the search results and whether the number is sufficient, but also whether the content of the object of interest in the search results is Front.

In this embodiment, the calculation method of word-of-mouth reputation includes: calculating the word-of-mouth reputation of the object of interest obtained based on the direct display content of search result entries associated with the object of interest in the set, and / or the calculation is based on and The word-of-mouth reputation of the attention object obtained by indirectly displaying the content of the search result item associated with the attention object in the collection.

It should be noted that the title and abstract of the search results obtained based on the website search are usually displayed directly on the search page, and the content of the manuscripts of the search results can only be obtained after clicking the link to enter. Therefore, the manner of calculating the reputation of the concerned object obtained based on the direct display of the content of the search result entry associated with the concerned object in the collection may be analyzed based on the title and summary of the search result, or based on the search Analysis of the results of the manuscript content, of course, comprehensive analysis based on the title, abstract and manuscript content of the search results.

In one embodiment, the calculation is based on the popularity of the object of interest obtained based on the direct display of the content of the search result entries associated with the object of interest in the collection, and the steps shown in FIG. 2 are specifically performed:

S21: Perform sentiment analysis on the context in which the subject of interest is in the title and summary of each search result entry, and determine the word of mouth of the subject of interest based on each search result entry based on the sentiment analysis result.

S22: According to the closeness of the sentiment analysis result of each search result item to the positive emotion, a search result item that meets the requirements of word of mouth is screened.

S23: Calculate the ratio of the number of search result entries that meet the word-of-mouth requirements to the number of search result entries associated with the object of interest, and judge the word-of-mouth degree of the object of interest based on the calculated proportion.

Optionally, the sentiment analysis result may be specifically divided into a positive evaluation, a neutral evaluation, and a negative evaluation. Of course, according to different application scenarios, the sentiment analysis results may have different classification standards, so as to obtain different classification categories and quantities, which are not limited in the present invention.

Taking a specific application scenario as an example, the concerned object is "A brand", and the search term is "Which is the strongest sweeping robot"? Based on the search term, a search is performed on the B site to obtain multi-page search results. 3 pages as a collection of search results. It should be noted that different pages may be selected as the search result set in different embodiments, which is not limited to this embodiment. In the search result set, search results that match "A brand" are filtered. For example, the title or abstract includes "A brand", or the title summary includes "A brand" and other criteria for filtering. Perform a sentiment analysis on the context of "Brand" in the title and summary of each search result item that has been filtered out, and assign 3 points to the positively rated search result items and 2 points to the neutrally rated search result items according to the sentiment analysis results. A negative evaluation search result entry is given 1 point, and a point value of 2 points or more is defined as meeting word of mouth requirements. Therefore, the number of all search result entries with a score of 2 and 3 can be counted, and the proportion of this number to the total number of all search result entries that are filtered out can be used as the reputation of "A brand".

It should be noted that the sentiment analysis in this embodiment can be implemented by a semantic analysis algorithm, such as the NLP natural language processing algorithm, which uses speculation, probability, statistics and other methods to perform sentiment analysis based on the context to determine whether the title and abstract are relevant to the Whether the concerned object is a positive evaluation, a negative evaluation, or a neutral evaluation, and then, for example, the sentiment dictionary and Bayesian algorithm can be used to classify the sentiment of the text. The above semantic analysis algorithms are all existing technologies, so they will not be described again.

The calculation is based on the reputation of the attention object obtained based on the indirect display of search result entries associated with the attention object in the collection, and the calculation method specifically includes

In an embodiment, the calculating the word-of-mouth reputation of the concerned object obtained based on the indirect display of content of the search result entries associated with the concerned object in the set, which specifically performs the steps shown in FIG. 3 :

S31: Determine the positions of the object of interest in the document content of the search result entry.

S32: Perform an sentiment analysis on the context in which the subject of interest is in each of the positions and obtain sentiment analysis results of each position, and comprehensively analyze the sentiment analysis of the search result entry according to the sentiment analysis results corresponding to each position. result.

S33: According to the closeness of the sentiment analysis result of each search result item to the positive emotion, a corresponding weight is given to each search result item.

S34: Filter out search result entries whose weights meet word of mouth requirements, and calculate the proportion of the number of these search result entries to the total number of search result entries associated with the object of interest. The calculated proportion is used to determine Describe the word of mouth of the object of interest.

Because the content of the manuscript appears repeatedly in the content of the manuscript, for example, a topic in a forum contains many netizens' evaluations of the people who are concerned about the appraisal. These reviews have positive, neutral, and negative evaluations. The comprehensive evaluation is a technical problem that needs to be solved in this embodiment. Therefore, the method of analyzing the word-of-mouth reputation of the concerned object based on the content of the manuscript is different from the method of analyzing the word-of-mouth reputation of the concerned object based on the title and summary.

Specifically, the sentiment analysis results of each location include non-negative evaluation and negative evaluation, and the non-negative evaluation includes positive evaluation and neutral evaluation; where: if the sentiment analysis results of all locations are non-negative evaluation, determine the The sentiment analysis result of the search result item is good and the search result item is given a high weight, for example, 5 points; if the number of positions where the sentiment analysis result is non-negative evaluation is greater than the number of positions where the sentiment analysis result is negative evaluation, Then determine that the sentiment analysis result of the search result item is better and give the search result item a higher weight, such as 4 points; if the sentiment analysis result is a non-negative evaluation, the number of positions equal to the sentiment analysis result is a negative evaluation The number of positions determines that the sentiment analysis result of the search result item is medium and gives the search result item a medium weight, such as 3 points; if the sentiment analysis result is a non-negative evaluation, the number of positions is less than the sentiment analysis result is negative Number of evaluated locations to determine sentiment analysis results for this search result entry Poor and assign a lower weight to the search result entry, for example, 2 points; if the sentiment analysis results of all locations are negative evaluations, determine that the sentiment analysis result of the search result entry is poor and assign the search result entry A low weight, such as 1 point.

In this embodiment, the weight value of 3 points and above is defined as meeting the requirements of word of mouth, so the word of mouth can be calculated according to the following formula:

It should be noted that the classification and weight assignment of the sentiment analysis results of the search result entries in this embodiment are for reference only, and are not used to limit the technical solution of the present invention.

In an embodiment, a manner of analyzing the visibility of search result entries associated with the object of interest in the search result entry set includes:

Among them, pct _i represents the visibility of the i-th search result entry in the search result entry set, x _i represents the assignment of the i-th search result entry in the page, and n represents n search result entries; let a preset number of searches The result entries are classified into the same search page, and the weighted visibility of each search result entry is calculated according to the weight of the search page in all search pages.

among them,

weight _j represents the weight of the j-th search page among all search pages, and x _ij represents the assignment of the i-th search result entry in the j-th search page.

It should be noted that, for the assignment of x _i , a subjective method or an objective method may be adopted, which is not limited in the present invention. The subjective valuation method refers to the raw data used to calculate the weight, which is mainly obtained by the assessor based on subjective judgments based on experience, such as subjective weighting, expert investigation, analytic hierarchy process, comparative weighting, multivariate analysis, and fuzzy statistics. The objective assignment method refers to the calculation of weighted raw data obtained from the actual data of the evaluation index during the evaluation process. For example, the variance method, principal component analysis method, entropy method, CRITIC method, etc. can be used, because the above methods themselves are Existing, so I will not repeat them.

In an embodiment, the influence evaluation parameter of the concerned object further includes a concentration, and the concentration is used to represent a proportion of search result entries associated with the concerned object in the set, and the calculation method of the concentration includes :

Among them, y ₁ represents search result entries associated with the object of interest, Counts of y ₁ represents the number of search result entries associated with the object of interest; Total y represents the total number of search result entries in the set.

It should be noted that word-of-mouth, visibility, concentration, and heat values are four different ways to describe the impact strength of the object of interest, which are collected from the praise degree of the object of interest and the search result entries associated with the object of interest in the collection The four dimensions of the degree of ranking before and after, and the number of search result entries associated with the object of interest in the collection, the search volume of the search term, or the degree of attention or popularity value, analyze the influence strength of the object of interest. Word-of-mouth, visibility, concentration, and heat value can be used alone or in combination of two or three parameters, and can be used together for comprehensive analysis. The following specific embodiments are taken as examples to facilitate further understanding by those skilled in the art. Technical solution of the present invention.

In this embodiment, the user provides information such as a search term, an object of interest, a preset website with a search function, and an official website address. The impact detection method of the technical solution of the present invention specifically performs the following steps to determine the impact strength of the object of interest.

Step 1: Import each piece of information provided by the user into a system for performing an impact detection method.

Step 2: The web crawler crawls the first n pages or n search result entries obtained by searching based on the search term on the preset website with a search function. Among them, each search result entry obtained through a web crawler mainly includes two pieces of content, which are the titles and abstracts that can be directly observed on the search page, and the ones that can be obtained only after clicking the title or summary or the relevant link of the search result entry Manuscript content.

Step 3: Calculate the visibility of each search result item according to the visibility calculation formula above.

Step four: filter out search result entries associated with the object of interest and calculate the sum of the visibility of the search result entries as the visibility value of the object of interest.

Step 5: Based on the semantic analysis algorithm, analyze the subject of interest in the title and summary of the search result entry as positive, neutral, and negative evaluation, so as to obtain the reputation of the subject of interest based on the title and summary of the search result entry.

Step 6. Click the title or abstract or related link to enter the content of the search result entry, and perform sentence segmentation according to the location of the object of interest. Each sentence after the disconnection is used as the object to be tested. Based on the semantic analysis algorithm, analyze whether the sentence in which the subject of interest belongs is a non-negative evaluation or a negative evaluation, and comprehensively analyze the sentiment of the content of the manuscript according to the number of non-negative evaluations and negative evaluations. The scores are distributed from 1 to 5 points.

Step 7. Calculate the score of each search result entry according to the method of step 6, and filter out the search result entries that meet the requirements of word of mouth according to the size of the score. Calculate the ratio of the number of search result entries that meet the word-of-mouth requirements to the total number of all search result entries that match the object of interest, so as to obtain the word-of-mouth degree obtained by the object of interest based on the content of the manuscript of the search result entry.

Step 8. Repeat step 1 to step 7 within a preset period of time, and compare the word-of-mouth reputation and visibility calculated each time with the previous or previous calculation results to obtain the word-of-mouth trends and visibility trends. . If the popularity is increasing, it means that the praise level of the concerned object is increasing, otherwise it means that the praise level of the concerned object is declining; if the visibility is increasing, it means that the attention level is getting better and better. On the contrary, it indicates that the ranking noticed by the subject is getting worse and worse.

In one embodiment, the influence detection method further analyzes a user's performance on the platform he is following and the relationship between the user and a competitor by calculating a comprehensive influence index of word of mouth. The calculation method of the word-of-mouth influence comprehensive index specifically executes each step shown in FIG. 4.

S41: Calculate the reputation of the same subject of interest based on a plurality of different search terms based on a preset website.

S42: Calculate the website word-of-mouth index of the concerned object based on the website based on the word-of-mouth reputation obtained by a predetermined website based on multiple different search terms and the popularity of each search term on the website. The calculation formula of the website word of mouth index is:

Among them, pct _p1 , ..., pct _pn indicate that the object of interest is based on the word of mouth obtained by the website based on the 1st to nth search terms, and V1, ..., Vn indicate the popularity value of the 1st to n search terms on the website . It should be noted that the specific form of the heat value varies with different network platforms, and may change with the development of the network platform. For example, the popularity value of Baidu's website mainly refers to the search volume; the popularity value of today's headline website mainly refers to the popularity index, which weights and sums the user's reading, analysis, or commentary behavior according to the popularity index model and obtains the corresponding The heat index of an event, article, or keyword is usually drawn as a trend chart in units of hours or days to show the change of the heat index with the event. Knowing the heat value of the website mainly refers to the number of topics followed. WeChat website The popularity value mainly refers to the WeChat index and so on. Due to the different statistical methods of various websites, the representation value of the popularity value is also different, and this embodiment will not list them one by one.

S43: Calculate and obtain the word-of-mouth influence index of the object of interest based on the website according to the word-of-mouth reputation index of the website and the usage rate of the website within a preset time period. The calculation formula of the word-of-mouth influence index is:

Among them, Web_Mount Percent indicates the usage rate of the website within a preset time period.

S44: Add the word-of-mouth influence index of each website on the network to generate a corresponding word-of-mouth influence comprehensive index; the word-of-mouth influence comprehensive index is used to indicate the word-of-mouth performance of the concerned object on the entire network. The calculation formula of the word-of-mouth influence comprehensive index is: Total_Web_Index _p = ΣWeb_Index _p . It should be noted that, because the word-of-mouth influence comprehensive index reflects the pros and cons between the customer and the competing brand, which is closely related to the needs of consumers, the search term is preferably a consumer demand term.

In a specific application scenario, "A Airlines" is taken as the object of interest in this embodiment, Baidu website is used as the preset website in this embodiment, and "Special Airfare", "Online Check-in", Terms such as "flight delay", "airline recruitment", and "airplane meal" are used as search terms in this embodiment. The word-of-mouth reputation of "A Airlines" on Baidu's website can be obtained by calculating the word-of-mouth reputation of Baidu's website according to the search terms. In this embodiment, it is assumed that "A airline"'s word-of-mouth reputation on Baidu's website As shown in Figure 5, the degrees are 60%, 50%, 46%, 24%, 38% ... according to the following formula

Can obtain "A Airlines" website reputation index based on Baidu website. By using the same calculation principle, it is also possible to calculate the website word-of-mouth index of A airline's competitors based on Baidu website, and also calculate the website word-of-mouth index of airline A based on other websites.

For ease of description, this embodiment assumes that the “A Airlines” website reputation index based on Baidu website is 20%, the website reputation index based on Zhihu website is 15%, and the website reputation index based on Taobao website is 5%. In addition, based on statistics or data provided by third parties, assuming that the number of visitors to each site in the month is 20,000, 30,000, and 40,000, and assuming that the total number of Internet users on the network is 200,000, the monthly usage rate of each site is approximately It is 20000/200000, 30000/200000, 40000/200000, that is, 0.1, 0.15, 0.2 respectively. According to formula

The word-of-mouth influence index of each website can be calculated, that is, the word-of-mouth influence index of Baidu website, Zhihu website and Taobao website are 20% * 0.1, 15% * 0.15, 5% * 0.2, respectively. According to the word-of-mouth influence index, the performance of the concerned object on each website and the performance of the customer and competitors on the same website can be judged, thereby helping the customer to make targeted comparisons with competitors.

The word-of-mouth influence index of each website is calculated and summed to obtain the comprehensive word-of-mouth influence index, Total_Web_Index _p = ΣWeb_Index _p . The word-of-mouth influence comprehensive index represents the word-of-mouth index of the object of interest based on the entire network. The word-of-mouth index based on the entire network is different for different objects of interest. Therefore, the performance of the object of interest and its opponents based on the entire network can be judged based on the size of the index.

It should be noted that the above calculation methods for the website word of mouth index, word of mouth influence index, and word of mouth influence comprehensive index can also be applied to the calculation of visibility, that is, the website visibility and website visibility index are calculated based on the same calculation principle. Website visibility and website visibility index are also used for customers to know the pros and cons of their relationship with competitors.

Specifically, the calculation formula of the website visibility index is:

Among them, pct _s1 , ..., pct _sn indicate the visibility of the object of interest based on the website's first to nth search terms, and V1, ..., Vn indicate the popularity value of the first to n search terms on the website; The type of the heat value includes any one or more combinations of a search amount, an attention amount, and a heat index. The calculation formula of the visibility influence index is: Web_Index _s = Web_pct _s * Web_Mount Percent; where Web_Mount Percent represents the usage rate of the website within a preset time period. The calculation formula of the visible influence comprehensive index is: Total_Web_Index _s = ΣWeb_Index _s . Because the calculation process is similar to the calculation process of the word-of-mouth influence composite index above, it will not be repeated here.

As shown in FIG. 6, a two-dimensional matrix analysis diagram based on search volume and visibility in an embodiment of the present invention is shown. This embodiment is divided into 4 matrices according to the search volume and visibility, still using "A Airlines" as an example: "A Airlines" calculates the visibility and search volume based on the consumer keywords that fall into the matrix in the lower left corner. Poor, so it is not recommended to continue to promote consumer keywords in this matrix area on the Internet; "A Airlines" has better visibility and search volume calculated based on consumer keywords that fall into the upper right corner of the matrix, so It is recommended to maintain the current network promotion; "A Airlines" calculated based on consumer keywords that fall into the upper left corner matrix has better visibility but poor search volume, so it is recommended that keywords in this matrix area need to increase public relations. Cooperate to increase the search volume of keywords; "A Airlines" calculates the search volume calculated based on consumer keywords that fall into the matrix in the lower right corner, but the search volume is better but the visibility is poor, and customers may be advised to increase the keywords in the matrix area To promote visibility in high search volume.

As shown in FIG. 7, a two-dimensional matrix analysis diagram based on search volume and visibility comparison values in an embodiment of the present invention is shown. In this embodiment, the horizontal axis represents search volume, and the visibility of the vertical axis is formulated based on a comparison result of the visibility of an object of interest with a competitor. The upward ordinate indicates that the visibility of the object of interest is better than that of the competitor. Visibility, otherwise it lags behind competitors.

Among them, "A Airlines" calculates the visibility behind the search terms that fall into the matrix in the lower left corner behind the opponent and the search volume is poor, so it is not recommended to continue to promote consumer keywords in this matrix area on the Internet; "A "Airlines" calculated according to the search terms that fall into the upper right corner matrix have better visibility than their opponents and higher search volume. It is recommended to monitor the dynamic performance of search terms in this matrix area to prevent them from being surpassed; "A Airlines According to the search terms that fall into the upper left corner of the matrix, the calculated visibility is better than the opponent but the search volume is poor. Therefore, the search terms in this matrix area can be expanded to fully cover search terms that are meaningful to consumers. Although the amount is not high, the total search volume is still more objective; the visibility calculated by "A Airlines" based on the search terms that fall into the lower right corner matrix lags behind the opponent but has a high search volume. It may be suggested that the performance of this matrix area must be improved.

Therefore, the two-dimensional matrix analysis chart provided in this embodiment can clearly show the user's own performance and relationship with competitors, and provide users with corresponding placement strategies according to different matrix areas. Which search terms are effective and worthy of placement? It ’s not worth the effort, it ’s clear and very efficient and accurate. In addition, the technical solution of the present invention also includes understanding the dynamic performance of each search term over a period of time through monitoring within a preset period of time, thereby helping the user to adjust the investment direction and budget plan in a timely manner.

As shown in FIG. 8, a two-dimensional matrix analysis diagram based on visibility and word of mouth in an embodiment of the present invention is shown. In this embodiment, the visibility-word-of-mouth reputation matrix is divided into 4 matrix blocks, and the visibility object and the word-of-mouth reputation calculated by the search terms that fall into the upper right corner matrix are both excellent, so it is recommended to keep the current Word-of-mouth reputation and visibility level; the object of interest calculated based on search terms that fall into the upper left corner matrix has better word-of-mouth reputation but poor visibility, so it is recommended to increase visibility in a targeted manner while maintaining word-of-mouth significance. ; The word of interest and visibility obtained from the search terms that fall into the lower left corner matrix are poor, so it is recommended to improve the word of mouth and visibility; the object of interest is based on the search words that fall into the lower right corner matrix; The calculated visibility is better but word-of-mouth is poor, so it is recommended to improve the word-of-mouth in a targeted manner while maintaining positional significance as much as possible. The two-dimensional matrix provided in this embodiment can help users clearly understand the current word-of-mouth and visibility situation, thereby helping users adjust investment strategies in a timely manner.

As shown in FIG. 9, a multi-dimensional matrix analysis diagram based on search volume, visibility, and word of mouth in an embodiment of the present invention is shown. In this embodiment, the horizontal axis represents the search volume, the vertical axis represents the visibility, and the size of the circle area indicates the level of word of mouth. For example: "Airline A" calculates a higher search volume based on the search term "special airfare", has lower visibility and poor word of mouth, and calculates a lower search volume based on the search term "A airline stock", resulting in lower visibility Low but good word of mouth. The search volume calculated based on the search term "cheap airfare" has average visibility and word of mouth, while the search volume calculated based on the search term "economic seat" is poor, with high visibility and good word of mouth. .

Therefore, the multi-dimensional matrix analysis chart provided in this embodiment can directly and directly obtain the placement effect of each search term through the heat value, visibility, and word of mouth, so that the factors that affect the influence strength of the object of interest can be more thoroughly analyzed because of visibility. Insufficient or insufficient search volume, or insufficient word-of-mouth of the content of the manuscript, thereby helping users to make targeted improvements and adjust the impact strength of the target of attention.

Those of ordinary skill in the art may understand that all or part of the steps of implementing the foregoing method embodiments may be completed by computer program related hardware. The aforementioned computer program may be stored in a computer-readable storage medium. When the program is executed, the steps including the foregoing method embodiments are executed; and the foregoing storage medium includes: various media that can store program codes, such as a ROM, a RAM, a magnetic disk, or an optical disc.

As shown in FIG. 10, a schematic structural diagram of an electronic terminal according to an embodiment of the present invention is shown. The electronic terminal provided in this example includes a processor 101, a memory 102, a transceiver 103, a communication interface 104, and a system bus 105; the memory 102 and the communication interface 104 are connected to the processor 101 and the transceiver 103 through the system bus and complete each other. In the communication, the memory 102 is used to store a computer program, the communication interface 104 and the transceiver 103 are used to communicate with other devices, and the processor 101 is used to run the computer program, so that the electronic terminal performs the steps of the influence detection method as described above.

The system bus mentioned above may be a Peripheral Component Interconnect Standard (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The system bus can be divided into address bus, data bus, control bus and so on. For ease of representation, the figure only uses a thick line to represent, but it does not mean that there is only one bus or one type of bus. The communication interface is used to implement communication between the database access device and other devices (such as a client, a read-write library, and a read-only library). The memory may include random access memory (RandomAccessMemory, RAM for short), and may also include non-volatile memory (non-volatile memory), such as at least one disk memory.

The above-mentioned processor may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (Network Processor, NP), etc .; it may also be a Digital Signal Processor (DigitalSignalProcessing, DSP), an application specific integrated circuit ( ApplicationSpecificIntegratedCircuit (ASIC for short), Field-ProgrammableGateArray (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

In summary, the present invention provides an impact detection method, an electronic terminal, and a storage medium that are applicable to the object of interest, and has the following beneficial effects: The present invention is based on a number of impact evaluation parameter analyses such as word of mouth, visibility, concentration, and heat value This article describes the impact strength of the concerned objects, thereby effectively and comprehensively evaluating the brand's network influence. Therefore, the present invention effectively overcomes various shortcomings in the prior art and has high industrial utilization value.

The above-mentioned embodiments merely illustrate the principle of the present invention and its effects, but are not intended to limit the present invention. Anyone familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or changes made by those with ordinary knowledge in the technical field to which they belong without departing from the spirit and technical ideas disclosed by the present invention should still be covered by the claims of the present invention.

Claims (15)

1. An impact detection method suitable for an object of interest, which includes:

   Use search terms to get a collection of search result entries;

   Calculating an influence evaluation parameter of the attention object obtained based on search result entries associated with the attention object in the set, which includes word-of-mouth degree and / or visibility; wherein the word-of-mouth degree representation is obtained based on the set The degree of praise of the concerned object; the visibility indicates the degree of ranking of the search result entries associated with the concerned object in the set.
2. The method for detecting influence applicable to an object of interest according to claim 1, wherein the object of interest includes any one or more of a brand name, a product name, a company CEO name, a company slogan, or a competing product Combination; the search term includes any one or more of a consumer demand word, a brand word, a brand extension word, a company word, and a competition word.
3. The method for detecting influences applicable to an object of interest according to claim 1, wherein the calculation method of the word of mouth comprises:

   Calculate the word-of-mouth reputation of the attention object obtained based on the direct display content of search result entries associated with the object of interest in the set, and / or calculate the reputation of the object of interest based on search result entries associated with the object of interest in the set The reputation of the object of interest obtained by indirect display of content; wherein the direct display content includes the title and summary of the search result entry, and the indirect display content includes the content of the manuscript of the search result entry.
4. The method for detecting influence of an object of interest according to claim 3, wherein the calculation is based on the object of interest obtained by directly displaying content of a search result entry associated with the object of interest in the set The calculation method of word of mouth includes:

   Perform sentiment analysis on the context in which the subject of interest is in the title and summary of each search result entry, and judge the word of mouth of the subject of interest based on each search result entry according to the sentiment analysis result;

   According to the closeness of the sentiment analysis results of each search result item to the positive emotions, filter out search result items that meet the word-of-mouth requirements;

   Calculate the ratio of the number of search result entries that meet the requirements for word-of-mouth to the total number of search result entries associated with the object of interest, and determine the word-of-mouth degree of the object of interest based on the calculated proportion.
5. The method for detecting influence of an object of interest according to claim 3, wherein the calculation is based on the object of interest obtained based on indirect display of search result entries associated with the object of interest in the collection. Of word-of-mouth, its calculation methods include:

   Determining the positions of the object of interest in the content of the manuscript of the search result entry;

   Performing an sentiment analysis on the context in which the subject of interest is in each of the positions to obtain sentiment analysis results of the respective positions, and comprehensively obtained the sentiment analysis results of the search result entry according to the sentiment analysis results corresponding to each position;

   Assign a corresponding weight value to each search result item according to the closeness of the sentiment analysis result of each search result item to the positive emotion;

   Filter out search result entries whose weight meets word of mouth requirements, and calculate the proportion of the number of these search result entries to the total number of search result entries associated with the object of interest. The calculated proportion is used to determine the attention Object's reputation.
6. The method for detecting influences applicable to a subject of interest according to claim 5, wherein the sentiment analysis results of each location include non-negative evaluation and negative evaluation, and the non-negative evaluation includes positive evaluation and neutral evaluation;

   among them:

   If the sentiment analysis results of all positions are non-negative evaluations, determine that the sentiment analysis result of the search result item is good and give the search result item a high weight;

   If the number of positions where the sentiment analysis result is non-negative evaluation is greater than the number of positions where the sentiment analysis result is negative evaluation, determine that the sentiment analysis result of the search result item is better and assign a higher weight to the search result item;

   If the number of positions where the sentiment analysis result is a non-negative evaluation is equal to the number of positions where the sentiment analysis result is a negative evaluation, determining that the sentiment analysis result of the search result item is medium and assigning the search result item a medium weight;

   If the number of positions where the sentiment analysis result is non-negative evaluation is less than the number of positions where the sentiment analysis result is negative evaluation, determine that the sentiment analysis result of the search result item is poor and assign a lower weight to the search result item;

   If the sentiment analysis results of all positions are negative evaluations, it is determined that the sentiment analysis result of the search result item is poor and the search result item is given a low weight.
7. The method for detecting influence of a subject of interest according to claim 6, wherein the word-of-mouth degree is:

   
8. The method for detecting influence of a subject of interest according to claim 1, wherein the manner of analyzing the visibility of search result entries associated with the subject of interest in the set of search result entries comprises:

   

   Among them, pct _i represents the visibility of the ith search result entry in the search result entry set, x _i represents the assignment of the ith search result entry in the page, and n represents n search result entries;

   Make a preset number of search result entries into the same search page, and calculate the weighted visibility of each search result entry according to the weight of the search page in all search pages,

   

   among them,

   

   weight _j represents the weight of the j-th search page among all search pages, and x _ij represents the assignment of the i-th search result entry in the j-th search page.
9. The method for detecting influence of a subject of interest according to claim 1, wherein the influence evaluation parameter of the subject of interest further comprises a concentration, and the concentration is used to represent a search result entry associated with the subject of interest The proportion of the concentration in the set, the calculation method of the concentration includes:

   

   Among them, y ₁ represents search result entries associated with the object of interest, Counts of y ₁ represents the number of search result entries associated with the object of interest; Total y represents the total number of search result entries in the set.
10. The method for detecting influences applicable to an object of interest according to claim 1, comprising:

    Calculate the reputation of the same object of interest based on multiple different search terms based on a preset website;

    Calculating the word-of-mouth reputation index of the concerned object based on the website based on the word-of-mouth reputation obtained by a preset website according to a plurality of different search terms and the popularity value of each search term on the website according to the concerned object;

    Based on the website word of mouth index and the usage rate of the website within a preset time period, calculating the object of interest based on the website word of mouth influence index;

    The word-of-mouth influence index of each website on the network is added up to generate a corresponding comprehensive index of word-of-mouth influence; the comprehensive index of word-of-mouth influence is used to indicate the word-of-mouth performance of the object of interest on the entire network.
11. The method for detecting influences applicable to an object of interest according to claim 10, further comprising:

    The calculation formula of the website word of mouth index is:

    

    Among them, pct _p1 , ..., pct _pn indicate that the object of interest is based on the word of mouth obtained by the website based on the 1st to nth search terms, and V1, ..., Vn indicate the popularity value of the 1st to n search terms on the website ; Wherein the type of the heat value includes any one or more combinations of a search amount, a focus amount, and a heat index.

    The calculation formula of the word-of-mouth influence index is:

    

    Among them, Web_Mount Percent indicates the usage rate of the website within a preset time period;

    The calculation formula of the word-of-mouth influence comprehensive index is: Total_Web_Index _p = ΣWeb_Index _p .
12. The method for detecting influences applicable to an object of interest according to claim 1, comprising:

    Calculate the visibility of the same subject of interest based on multiple different search terms based on a preset website;

    Calculating the visibility index of the object of interest based on the website based on the visibility obtained by a preset website based on multiple different search terms and the popularity value of each search term on the website;

    Based on the visibility index of the website and the usage rate of the website within a preset period of time, based on the visibility impact index of the website, the calculated object of interest is calculated;

    The visibility impact index of each website on the network is summed to generate a corresponding visibility impact comprehensive index; the visibility impact comprehensive index is used to indicate the ranking performance of the object of interest on the entire network .
13. The method for detecting influences applicable to an object of interest according to claim 12, further comprising:

    The calculation formula of the website visibility index is:

    

    Among them, pct _s1 , ..., pct _sn indicate the visibility of the object of interest based on the website's first to nth search terms, and V1, ..., Vn indicate the popularity value of the first to n search terms on the website; The type of the heat value includes any one or more combinations of search volume, attention volume, and heat index;

    The calculation formula of the visibility influence index is Web_Index _s = Web_pct _s * Web_Mount Percent; wherein Web_Mount Percent represents the usage rate of the website within a preset time period;

    The calculation formula of the visible influence comprehensive index is: Total_Web_Index _s = ΣWeb_Index _s .
14. A computer-readable storage medium having stored thereon a computer program, characterized in that when the program is executed by a processor, the method for detecting an influence force applicable to an object of interest according to any one of claims 1 to 13 is implemented.
15. An electronic terminal, comprising: a processor and a memory;

    The memory is configured to store a computer program, and the processor is configured to execute the computer program stored in the memory, so that the terminal performs the influence detection according to any one of claims 1 to 13, which is applicable to an object of interest. method.

Images