Information processing method, processing device, electronic equipment and readable storage medium
By acquiring text data and using classification models and popularity values to calculate recommendation indices, the problem of inaccurate information source recommendations in existing technologies is solved, enabling reliable information source recommendations and diversified utilization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NAT COMP NETWORK & INFORMATION SECURITY MANAGEMENT CENT
- Filing Date
- 2021-04-21
- Publication Date
- 2026-07-10
AI Technical Summary
Existing technologies lack effective means to reliably recommend information sources to users and ensure the value of these recommendations in business applications.
By acquiring text data, the relevance of the text data is determined using the classification model of the target business, and the recommendation index is calculated by combining the popularity value of the target Uniform Resource Locator (URL). The target URL is then output to achieve the recommendation of information sources.
It achieves accurate recommendation of information sources, ensuring a strong correlation between the recommended information sources and the target business, thereby improving the credibility and diversified utilization of the recommendations.
Smart Images
Figure CN115310000B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of information recommendation technology, and more specifically, to an information processing method, processing device, electronic device, and readable storage medium. Background Technology
[0002] Existing contacts send the page they want to share to the users they want to share with. Specifically, they send the webpage link of the page they want to share with directly to the users they want to share with, so that the users they want to share with can open the link and view it.
[0003] The inventors of this invention have discovered that existing sharing methods have limited value, and no one in the art has yet proposed an effective technical means to reliably recommend information sources to users and ensure the utilization value of such recommendations in business applications. Summary of the Invention
[0004] The present invention aims to solve, or at least partially solve, the technical problems existing in the prior art or related art.
[0005] In view of this, according to a first aspect of the present invention, the present invention provides an information processing method, comprising: acquiring text data; obtaining a target Uniform Resource Locator (URL) based on the text data; inputting the text data into a classification model of a target business to obtain a target business relevance of the text data, wherein the target business relevance is a value representing the degree of relevance between the text data and the target business; determining a popularity value for the target URL; determining a recommendation index for the target URL based on the popularity value and the target business relevance of the text data; and outputting the target URL based on the recommendation index.
[0006] The technical solution of this application proposes an information processing method. By running this method, recommendation indices can be labeled for target URIs extracted from text data, and target URIs can be output according to the recommendation indices to realize the recommendation of information sources and realize the diversified utilization of target URIs.
[0007] Specifically, text data is acquired, and a target Uniform Resource Locator (URL) is obtained from the text data. During this process, the data contained in the text data is organized to identify sites, social media accounts, etc. that frequently publish information relevant to the target business.
[0008] In one technical solution, the text data can be an article sent by a user that is related to the target business. The article can contain information such as information about leading figures in the industry or field in which the target business is located, research progress, and advertisements.
[0009] In one technical solution, text data is input into a classification model of the target business to obtain the relevance of the text data to the target business. In other words, the classification model of the target business is used to predict the category of the text data in order to determine the probability that the text data belongs to the target business.
[0010] In this technical solution, the classification model for the target business can be a pre-trained classification model. Specifically, it is a model trained using information that the target business is interested in and information that is not related to the target business.
[0011] In one of the technical solutions, the classification model for the target business is a binary classification model.
[0012] One technical solution includes cleaning the text data before inputting it into the target business classification model, such as removing target Uniform Resource Locators (URLs) and symbols like @ from the text data, in order to ensure the credibility of the target business relevance of the output text data.
[0013] In one technical solution, the popularity value of the target URI is determined. Specifically, this reflects the degree of diffusion of the target URI on different platforms and its occurrence on any platform. By statistically analyzing the popularity value of the target URI, recommendations can be made based on the popularity of the target URI when determining the recommendation index. This ensures that the recommended information sources are commonly used or frequently appearing sources, thus ensuring the credibility of the recommended information sources.
[0014] Furthermore, since the recommendation index is determined based on the popularity value and the relevance of the text data to the target business, it can ensure that the recommended information source has a strong correlation with the target business, thus achieving accurate information source recommendation.
[0015] In addition, the information processing method in the above-mentioned technical solution provided by the present invention may also have the following additional technical features:
[0016] In the above technical solution, obtaining text data and obtaining the target Uniform Resource Locator (URL) based on the text data specifically includes: extracting URLs from the text data to obtain the extraction result; determining the site type corresponding to the extraction result; and parsing the extraction result based on the parsing rules corresponding to the site type to obtain the target URL.
[0017] In this technical solution, by obtaining the site type corresponding to the extraction result, the extraction result can be parsed according to the parsing rules corresponding to the site type, which ensures the accuracy of the obtained target Uniform Resource Locator and reduces the occurrence of situations where the extraction result is not parsed correctly.
[0018] In one technical solution, for sites classified as social media sites, the target Uniform Resource Locator includes both the hostname domain and the path domain; for sites classified as non-social media sites, the target Uniform Resource Locator includes only the hostname domain.
[0019] In this technical solution, site types are divided into social media sites and non-social media sites. Under social media sites, the target uniform resource locator includes a hostname domain and a path domain. Since the target uniform resource locator also includes a path domain, it is possible to follow social media accounts, thus ensuring the credibility of the recommended information source.
[0020] Specifically, a Uniform Resource Locator typically consists of the following components in order: protocol field, hostname field, path field, filename field, and parameter field.
[0021] To ensure accurate parsing, the extracted results are categorized based on whether they belong to social media sites. If they belong to social media sites, the extracted results are parsed according to the parsing rules corresponding to social media sites to obtain the target Uniform Resource Locator (URL). If they do not belong to social media sites, the extracted results are parsed according to the parsing rules corresponding to non-social media sites to obtain the target URL.
[0022] Specifically, social media sites retain hostname and path domains. For example, a social media account like https: / / twitter.com / rxxxa / status / 1xxx7?s=21 resolves to...
[0023] https: / / twitter.com / rxxxa, where rxxxxa represents the username, which is the part corresponding to the path field.
[0024] Specifically, social media sites, which are essentially regular sites, retain the hostname domain. For example, https: / / www.kxxxr.com / xxx / mxxxy / lexxxa is re-resolved to https: / / www.kxxxr.com / .
[0025] One technical solution uses regular expressions to extract all Uniform Resource Locators (URLs) from the text data, thus obtaining the extraction results.
[0026] In one technical solution, a URL parser is used to parse the extraction results. The input of the URL parser is the extraction results, and the output is the website homepage, the homepage of social media people, organizations, etc.
[0027] One of the technical solutions involves obtaining text data and obtaining a target Uniform Resource Locator (URL) based on the text data. It also includes: simulating a request to the extracted results based on the fact that the extracted results are short links, in order to obtain the target URL.
[0028] In this technical solution, by determining whether the extracted result is a short link, a simulated request is made to the extracted result that is a short link if the determination result is yes, thereby obtaining the target Uniform Resource Locator (URL). This process expands the applicable scenarios for determining the target URL.
[0029] Specifically, given the word limits on content posted on social media platforms, longer links are usually shortened to a shorter link. Therefore, by determining whether the extracted result is a short link, the true Uniform Resource Locator can be displayed.
[0030] For example, the extracted short link https: / / t.co / 1cMxxxb?amp=1 is redirected to https: / / www.kxxxr.com / xxx / mxxxy / lexxxa.
[0031] In any of the above technical solutions, before determining the popularity value of the target Uniform Resource Locator (URL), the method further includes: comparing the target URL with a preset source database; determining whether the target URL does not exist in the preset source database; if so, storing the target URL in the preset source database.
[0032] In this technical solution, the target Uniform Resource Locator (URI) is compared with a preset source database to update the preset source database based on the comparison results. In this process, the discovery and updating of the target URI are realized, the preset source database is automatically maintained, and user involvement is reduced.
[0033] In any of the above technical solutions, the following method is also included: updating the recommendation index of the target Uniform Resource Locator based on the existence of the target Uniform Resource Locator in the preset source database.
[0034] In this technical solution, when a target Uniform Resource Locator (URL) is detected in the preset source database, the recommendation index of the target URL is updated to ensure that the recommendation index is up-to-date. This ensures that when recommending sources based on the recommendation index, sources that are suitable for the current situation can be recommended, thus ensuring the credibility of the recommended sources.
[0035] In any of the above technical solutions, the recommendation index of the target Uniform Resource Locator is determined based on the heat value and the target business relevance of the text data. Specifically, this includes: obtaining a first coefficient value and a second coefficient value; calculating the product of the first coefficient value and the heat value to obtain a first calculated value; calculating the product of the average relevance of the target business and the second coefficient value to obtain a second calculated value; and using the sum of the first calculated value and the second calculated value as the recommendation index of the target Uniform Resource Locator, wherein the sum of the first coefficient value and the second calculated value is 1.
[0036] In this technical solution, by introducing a first coefficient value and a second coefficient value, the sum of which is 1, the average value of the popularity value and the relevance of the target business can be adjusted when calculating the recommendation index of the target Uniform Resource Locator. This ensures that the calculated recommendation index accurately reflects the recommendability of the target Uniform Resource Locator and guarantees the credibility of the recommendation.
[0037] Specifically, rect k =β1hotness k +β2avg(p k，j ).
[0038] Where, rect k This represents the recommendation index of a target Uniform Resource Locator (URL) with a unique index k, where β1 is the first coefficient value, β2 is the second coefficient value, and avg(p) represents the recommendation index. k，j () represents the average relevance of the target business.
[0039] Where β1, β2∈[0,1], and β1+β2=1.
[0040] In any of the above technical solutions, the target Uniform Resource Locator (URL) is output based on the recommendation index, specifically including: determining the ranking of the recommendation index of the target URL in the recommendation set; and outputting the target URL based on the ranking.
[0041] In this technical solution, the recommendation indices in the recommendation set are sorted so that the output can be based on the sorting results. This ensures that the recommendation index of the target Uniform Resource Locator is relatively high, thus ensuring the credibility of the recommendation source.
[0042] In one of the technical solutions, the recommendation indices of the recommendation set are sorted from high to low.
[0043] In one of the technical solutions, before determining the popularity value of the target Uniform Resource Locator (URL), the following steps are included: obtaining a recommendation set from a preset source database; determining whether the target URL belongs to the recommendation set; and if so, calculating the popularity value of the target URL.
[0044] In this technical solution, by first determining whether the target Uniform Resource Locator (URL) belongs to the recommendation set, and then determining whether the popularity value of the target URL needs to be calculated based on the determination result, the calculation of the popularity value of the target URL is avoided when the target URL does not belong to the recommendation set. Thus, the total number of times the popularity value is calculated is reduced.
[0045] According to a second aspect of the present invention, an information processing apparatus is provided, comprising: an acquisition unit for acquiring text data and obtaining a target Uniform Resource Locator (URL) based on the text data; a first determination unit for inputting the text data into a classification model of a target business to obtain the target business relevance of the text data; a second determination unit for determining the popularity value of the target URL; a third determination unit for determining a recommendation index of the target URL based on the popularity value and the target business relevance of the text data; and an output unit for outputting the target URL based on the recommendation index.
[0046] The technical solution of this application proposes an information processing device, including an electronic device of the information processing device, which can label a recommendation index for target Uniform Resource Locators (URLs) extracted from text data, and output the target URLs according to the recommendation index, thereby realizing the diversified utilization of target URLs.
[0047] Specifically, text data is acquired, and a target Uniform Resource Locator (URL) is obtained from the text data. During this process, the data contained in the text data is organized to identify sites, social media accounts, etc. that frequently publish information relevant to the target business.
[0048] In one technical solution, the text data can be an article sent by a user that is related to the target business. The article can contain information such as information about leading figures in the industry or field in which the target business is located, research progress, and advertisements.
[0049] In one technical solution, text data is input into a classification model of the target business to obtain the relevance of the text data to the target business. In other words, the classification model of the target business is used to predict the category of the text data in order to determine the probability that the text data belongs to the target business.
[0050] In this technical solution, the classification model for the target business can be a pre-trained classification model. Specifically, it is a model trained using information that the target business is interested in and information that is not related to the target business.
[0051] In one of the technical solutions, the classification model for the target business is a binary classification model.
[0052] One technical solution includes cleaning the text data before inputting it into the target business classification model, such as removing target Uniform Resource Locators (URLs) and symbols like @ from the text data, in order to ensure the credibility of the target business relevance of the output text data.
[0053] In one technical solution, the popularity value of the target URI is determined. Specifically, this reflects the degree of diffusion of the target URI on different platforms and its occurrence on any platform. By statistically analyzing the popularity value of the target URI, recommendations can be made based on the popularity of the target URI when determining the recommendation index. This ensures that the recommended information sources are commonly used or frequently appearing sources, thus ensuring the credibility of the recommended information sources.
[0054] Furthermore, since the recommendation index is determined based on the popularity value and the relevance of the text data to the target business, it can ensure that the recommended information source has a strong correlation with the target business, thus achieving accurate information source recommendation.
[0055] In one of the technical solutions, the acquisition unit is specifically used to: extract Uniform Resource Locators (URLs) from text data to obtain extraction results; determine the site type corresponding to the extraction results; and parse the extraction results based on the parsing rules corresponding to the site type to obtain the target URL.
[0056] In this technical solution, by obtaining the site type corresponding to the extraction result, the extraction result can be parsed according to the parsing rules corresponding to the site type, which ensures the accuracy of the obtained target Uniform Resource Locator and reduces the occurrence of situations where the extraction result is not parsed correctly.
[0057] In one of the technical solutions, when the site type is a social media site, the target Uniform Resource Locator includes the hostname domain and the path domain; when the site type is a non-social media site, the target Uniform Resource Locator includes the hostname domain.
[0058] In this technical solution, site types are divided into social media sites and non-social media sites. Under social media sites, the target uniform resource locator includes a hostname domain and a path domain. Since the target uniform resource locator also includes a path domain, it is possible to follow social media accounts, thus ensuring the credibility of the recommended information source.
[0059] Specifically, a Uniform Resource Locator typically consists of the following components in order: protocol field, hostname field, path field, filename field, and parameter field.
[0060] To ensure accurate parsing, the extracted results are categorized based on whether they belong to social media sites. If they belong to social media sites, the extracted results are parsed according to the parsing rules corresponding to social media sites to obtain the target Uniform Resource Locator (URL). If they do not belong to social media sites, the extracted results are parsed according to the parsing rules corresponding to non-social media sites to obtain the target URL.
[0061] Specifically, social media sites retain hostname and path domains. For example, the social media account https: / / twitter.com / rxxxa / status / 1xxx7?s=21 resolves to https: / / twitter.com / rxxxa, where rxxxxa represents the username, which is the part corresponding to the path domain.
[0062] Specifically, social media sites, which are essentially regular sites, retain the hostname domain. For example, https: / / www.kxxxr.com / xxx / mxxxy / lexxxa is re-resolved to https: / / www.kxxxr.com / .
[0063] One technical solution uses regular expressions to extract all Uniform Resource Locators (URLs) from the text data, thus obtaining the extraction results.
[0064] In one technical solution, a URL parser is used to parse the extraction results. The input of the URL parser is the extraction results, and the output is the website homepage, the homepage of social media people, organizations, etc.
[0065] In one of the technical solutions, the acquisition unit is also used to: simulate a request to the extraction result based on the extraction result being a short link, in order to obtain the target Uniform Resource Locator.
[0066] In this technical solution, by determining whether the extracted result is a short link, a simulated request is made to the extracted result that is a short link if the determination result is yes, thereby obtaining the target Uniform Resource Locator (URL). This process expands the applicable scenarios for determining the target URL.
[0067] Specifically, given the word limits on content posted on social media platforms, longer links are usually shortened to a shorter link. Therefore, by determining whether the extracted result is a short link, the true Uniform Resource Locator can be displayed.
[0068] For example, the extracted short link https: / / t.co / 1cMxxxb?amp=1 is redirected to https: / / www.kxxxr.com / xxx / mxxxy / lexxxa.
[0069] In one of the technical solutions, the second determining unit is further configured to: compare the target Uniform Resource Locator (URI) with a preset source database; determine whether the target URI does not exist in the preset source database; if so, store the target URI in the preset source database.
[0070] In this technical solution, the target Uniform Resource Locator (URI) is compared with a preset source database to update the preset source database based on the comparison results. In this process, the discovery and updating of the target URI are realized, the preset source database is automatically maintained, and user involvement is reduced.
[0071] In one of the technical solutions, the third determining unit is also used to: update the recommendation index of the target Uniform Resource Locator based on the existence of the target Uniform Resource Locator in the preset source database.
[0072] In this technical solution, when a target Uniform Resource Locator (URL) is detected in the preset source database, the recommendation index of the target URL is updated to ensure that the recommendation index is up-to-date. This ensures that when recommending sources based on the recommendation index, sources that are suitable for the current situation can be recommended, thus ensuring the credibility of the recommended sources.
[0073] In one of the technical solutions, the third determining unit is specifically used to: obtain a first coefficient value and a second coefficient value; calculate the product of the first coefficient value and the popularity value to obtain a first calculated value; calculate the product of the average relevance of the target business and the second coefficient value to obtain a second calculated value; and use the sum of the first calculated value and the second calculated value as the recommendation index of the target unified resource locator, wherein the sum of the first coefficient value and the second coefficient value is 1.
[0074] In this technical solution, by introducing a first coefficient value and a second coefficient value, the sum of which is 1, the average value of the popularity value and the relevance of the target business can be adjusted when calculating the recommendation index of the target Uniform Resource Locator. This ensures that the calculated recommendation index accurately reflects the recommendability of the target Uniform Resource Locator and guarantees the credibility of the recommendation.
[0075] Specifically, rect k =β1hotness k +β2avg(p k，j ).
[0076] Where, rect k This represents the recommendation index of a target Uniform Resource Locator (URL) with a unique index k, where β1 is the first coefficient value, β2 is the second coefficient value, and avg(p) represents the recommendation index. k，j () represents the average relevance of the target business.
[0077] Where β1, β2∈[0,1], and β1+β2=1.
[0078] In one of the technical solutions, the output unit is specifically used to: determine the ranking of the recommendation index of the target Uniform Resource Locator in the recommendation set; and output the target Uniform Resource Locator based on the ranking.
[0079] In this technical solution, the recommendation indices in the recommendation set are sorted so that the output can be based on the sorting results. This ensures that the recommendation index of the target Uniform Resource Locator is relatively high, thus ensuring the credibility of the recommendation source.
[0080] In one of the technical solutions, the recommendation indices of the recommendation set are sorted from high to low.
[0081] In one of the technical solutions, the second determining unit is further configured to: obtain a recommendation set in a preset source database; determine whether the target Uniform Resource Locator belongs to the recommendation set; and if so, calculate the popularity value of the target Uniform Resource Locator.
[0082] In this technical solution, by first determining whether the target Uniform Resource Locator (URL) belongs to the recommendation set, and then determining whether to calculate the popularity value of the target URL based on the determination result, the calculation of the popularity value of the target URL is avoided when the target URL does not belong to the recommendation set, thereby reducing the total number of times the popularity value is calculated.
[0083] According to a third aspect of the present invention, an electronic device is provided, including a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the information processing method described above.
[0084] In this embodiment, the processor in the electronic device runs a program or instruction stored in the memory to implement the steps of any of the above-described information processing methods, thus possessing all the beneficial effects of any of the above-described information processing methods, which will not be elaborated further here.
[0085] According to a fourth aspect of the present invention, a readable storage medium is provided on which a program or instructions are stored, which, when executed by a processor, implement the steps of the information processing method as described above.
[0086] In this embodiment, when the program or instructions on the readable storage medium are executed, the steps of any of the above-described information processing methods are implemented. Therefore, the readable storage medium has all the beneficial effects of any of the above-described information processing methods, which will not be elaborated here.
[0087] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description
[0088] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:
[0089] Figure 1 A flowchart illustrating the information processing method in an embodiment of the present invention is shown;
[0090] Figure 2 This illustration shows a flowchart of the process of obtaining text data and obtaining a target Uniform Resource Locator (URL) based on the text data in an embodiment of the present invention.
[0091] Figure 3 This illustration shows a flowchart of maintaining a preset source database based on a target Uniform Resource Locator (URL) in an embodiment of the present invention.
[0092] Figure 4 This illustration shows a flowchart of a process for determining the recommendation index of a target Uniform Resource Locator (URL) based on the heat value and the target business relevance of text data in an embodiment of the present invention.
[0093] Figure 5 This illustration shows a flowchart of the process of outputting a target Uniform Resource Locator based on a recommendation index in an embodiment of the present invention.
[0094] Figure 6 A schematic block diagram of an information processing device according to an embodiment of the present invention is shown;
[0095] Figure 7 A schematic block diagram of an electronic device according to an embodiment of the present invention is shown;
[0096] Figure 8 A flowchart illustrating the control method of an electronic device according to an embodiment of the present invention is shown. Detailed Implementation
[0097] To better understand the above aspects, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.
[0098] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.
[0099] Example 1
[0100] like Figure 1 As shown, according to one embodiment of the present invention, the present invention provides an information processing method, comprising:
[0101] Step 102: Obtain text data and obtain the target Uniform Resource Locator (URL) based on the text data;
[0102] Step 104: Input the text data into the target business classification model to obtain the target business relevance of the text data;
[0103] Step 106: Determine the popularity value of the target Uniform Resource Locator;
[0104] Step 108: Determine the recommendation index of the target Uniform Resource Locator based on the popularity value and the target business relevance of the text data;
[0105] Step 110: Output the target Uniform Resource Locator based on the recommendation index.
[0106] The technical solution of this application proposes an information processing method. By running this method, recommendation indices can be labeled for target URIs extracted from text data, and target URIs can be output based on the recommendation indices, thereby realizing the diversified utilization of target URIs.
[0107] Specifically, text data is acquired, and a target Uniform Resource Locator (URL) is obtained from the text data. During this process, the data contained in the text data is organized to identify sites, social media accounts, etc. that frequently publish information relevant to the target business.
[0108] In one technical solution, the text data can be an article sent by a user that is related to the target business. The article can contain information such as information about leading figures in the industry or field in which the target business is located, research progress, and advertisements.
[0109] In one technical solution, text data is input into a classification model of the target business to obtain the relevance of the text data to the target business. In other words, the classification model of the target business is used to predict the category of the text data in order to determine the probability that the text data belongs to the target business.
[0110] In this technical solution, the classification model for the target business can be a pre-trained classification model. Specifically, it is a model trained using information that the target business is interested in and information that is not related to the target business.
[0111] In one of the technical solutions, the classification model for the target business is a binary classification model.
[0112] One technical solution includes cleaning the text data before inputting it into the target business classification model, such as removing target Uniform Resource Locators (URLs) and symbols like @ from the text data, in order to ensure the credibility of the target business relevance of the output text data.
[0113] In one technical solution, the popularity value of the target URI is determined. Specifically, this reflects the degree of diffusion of the target URI on different platforms and its occurrence on any platform. By statistically analyzing the popularity value of the target URI, recommendations can be made based on the popularity of the target URI when determining the recommendation index. This ensures that the recommended information sources are commonly used or frequently appearing sources, thus ensuring the credibility of the recommended information sources.
[0114] Furthermore, since the recommendation index is determined based on the popularity value and the relevance of the text data to the target business, it can ensure that the recommended information source has a strong correlation with the target business, thus achieving accurate information source recommendation.
[0115] Example 2
[0116] In this embodiment, the specific steps are defined as acquiring text data and obtaining a target Uniform Resource Locator (URL) based on the text data, such as... Figure 2 As shown, it includes:
[0117] Step 202: Extract Uniform Resource Locators (URLs) from the text data to obtain the extraction results;
[0118] Step 204: Determine the site type corresponding to the extraction results;
[0119] Step 206: Parse the extraction results based on the parsing rules corresponding to the site type to obtain the target Uniform Resource Locator.
[0120] In this technical solution, by obtaining the site type corresponding to the extraction result, the extraction result can be parsed according to the parsing rules corresponding to the site type, which ensures the accuracy of the obtained target Uniform Resource Locator and reduces the occurrence of situations where the extraction result is not parsed correctly.
[0121] Specifically, when the site type is a social media site, the target Uniform Resource Locator includes the hostname domain and the path domain; when the site type is a non-social media site, the target Uniform Resource Locator includes the hostname domain.
[0122] In this technical solution, site types are divided into social media sites and non-social media sites. Under social media sites, the target uniform resource locator includes a hostname domain and a path domain. Since the target uniform resource locator also includes a path domain, it is possible to follow social media accounts, thus ensuring the credibility of the recommended information source.
[0123] Specifically, a Uniform Resource Locator typically consists of the following components in order: protocol field, hostname field, path field, filename field, and parameter field.
[0124] To ensure accurate parsing, the extracted results are categorized based on whether they belong to social media sites. If they belong to social media sites, the extracted results are parsed according to the parsing rules corresponding to social media sites to obtain the target Uniform Resource Locator (URL). If they do not belong to social media sites, the extracted results are parsed according to the parsing rules corresponding to non-social media sites to obtain the target URL.
[0125] Specifically, social media sites retain hostname and path domains. For example, the social media account https: / / twitter.com / rxxxa / status / 1xxx7?s=21 resolves to https: / / twitter.com / rxxxa, where rxxxxa represents the username, which is the part corresponding to the path domain.
[0126] Specifically, social media sites, which are essentially regular sites, retain the hostname domain. For example, https: / / www.kxxxr.com / xxx / mxxxy / lexxxa is re-resolved to https: / / www.kxxxr.com / .
[0127] One technical solution uses regular expressions to extract all Uniform Resource Locators (URLs) from the text data, thus obtaining the extraction results.
[0128] In one technical solution, a URL parser is used to parse the extraction results. The input of the URL parser is the extraction results, and the output is the website homepage, the homepage of social media people, organizations, etc.
[0129] Example 3
[0130] In one embodiment, obtaining text data and obtaining a target Uniform Resource Locator (URL) based on the text data further includes: simulating a request to the extraction result based on the extraction result being a short link, in order to obtain the target URL.
[0131] In this technical solution, by determining whether the extracted result is a short link, a simulated request is made to the extracted result that is a short link if the determination result is yes, thereby obtaining the target Uniform Resource Locator (URL). This process expands the applicable scenarios for determining the target URL.
[0132] Specifically, given the word limits on content posted on social media platforms, longer links are usually shortened to a shorter link. Therefore, by determining whether the extracted result is a short link, the true Uniform Resource Locator can be displayed.
[0133] For example, the extracted short link https: / / t.co / 1cMxxxb?amp=1 is redirected to https: / / www.kxxxr.com / xxx / mxxxy / lexxxa.
[0134] Example 4
[0135] In this embodiment, a scheme for maintaining a preset source database based on a target Uniform Resource Locator (URL) is specifically defined, such as... Figure 3 As shown, it includes:
[0136] Step 302: Compare the target Uniform Resource Locator with the preset source database;
[0137] Step 304: Determine whether the target Uniform Resource Locator does not exist in the preset source database. If so, store the target Uniform Resource Locator in the preset source database.
[0138] In this technical solution, the target Uniform Resource Locator (URI) is compared with a preset source database to update the preset source database based on the comparison results. In this process, the discovery and updating of the target URI are realized, the preset source database is automatically maintained, and user involvement is reduced.
[0139] In this embodiment, the recommendation index of the target Uniform Resource Locator is updated based on the existence of the target Uniform Resource Locator in the preset source database.
[0140] In this technical solution, when a target Uniform Resource Locator (URL) is detected in the preset source database, the recommendation index of the target URL is updated to ensure that the recommendation index is up-to-date. This ensures that when recommending sources based on the recommendation index, sources that are suitable for the current situation can be recommended, thus ensuring the credibility of the recommended sources.
[0141] Example 5
[0142] In this embodiment, the recommendation index for the target Uniform Resource Locator (URL) is specifically defined based on the popularity value and the target business relevance of the text data. Figure 4 As shown, it includes:
[0143] Step 402: Obtain the first coefficient value and the second coefficient value;
[0144] Step 404: Calculate the product of the first coefficient value and the heat value to obtain the first calculated value;
[0145] Step 406: Calculate the product of the average relevance of the target business and the second coefficient value to obtain the second calculated value;
[0146] Step 408: Use the sum of the first calculated value and the second calculated value as the recommendation index for the target Uniform Resource Locator.
[0147] The sum of the first and second coefficients is 1.
[0148] In this technical solution, by introducing a first coefficient value and a second coefficient value, the sum of which is 1, the average value of the popularity value and the relevance of the target business can be adjusted when calculating the recommendation index of the target Uniform Resource Locator. This ensures that the calculated recommendation index accurately reflects the recommendability of the target Uniform Resource Locator and guarantees the credibility of the recommendation.
[0149] Specifically, rect k =β1hotness k +β2avg(p k，j ).
[0150] Where, rect k This represents the recommendation index of a target Uniform Resource Locator (URL) with a unique index k, where β1 is the first coefficient value, β2 is the second coefficient value, and avg(p) represents the recommendation index. k，j () represents the average relevance of the target business.
[0151] Where β1, β2∈[0,1], and β1+β2=1.
[0152] Example 6
[0153] In this embodiment, the output of the target Uniform Resource Locator (URL) is specifically defined based on the recommendation index, such as... Figure 5 As shown, it includes:
[0154] Step 502: Determine the ranking of the recommendation index of the target Uniform Resource Locator in the recommendation set;
[0155] Step 504: Output the target Uniform Resource Locator based on the sorting.
[0156] In this technical solution, the recommendation indices in the recommendation set are sorted so that the output can be based on the sorting results. This ensures that the recommendation index of the target Uniform Resource Locator is relatively high, thus ensuring the credibility of the recommendation source.
[0157] In one of the technical solutions, the recommendation indices of the recommendation set are sorted from high to low.
[0158] In one embodiment, before determining the popularity value of the target Uniform Resource Locator (URL), the method further includes: obtaining a recommendation set in a preset source library; determining whether the target URL belongs to the recommendation set; and if so, calculating the popularity value of the target URL.
[0159] In this technical solution, by first determining whether the target Uniform Resource Locator (URL) belongs to the recommendation set, and then determining whether to calculate the popularity value of the target URL based on the determination result, the calculation of the popularity value of the target URL is avoided when the target URL does not belong to the recommendation set. Thus, the total number of times the popularity value is calculated is reduced.
[0160] In one embodiment, the recommendation set is determined based on the target business relevance of text data; specifically, the text data includes URLs. k,j For example, the probability that it belongs to the target business is p. k,j Where k is a unique index identifier in the URL, j is a unique index identifier in the content source platform, and N k,j Indicates in the URL k The number of times p appears on platform j is determined by setting a threshold θ∈[0,1]. k,j >θ, the url k,j Add to the recommended collection.
[0161] The recommendation set is represented as follows:
[0162] [(url 1,1 N 1,1 ,p 1,1 ...(url) k,j N k,j ,p k,j...(url) K,J N K,J ,p K,J )).
[0163] In one embodiment, hotness k =α1∑ j sgn(N k,j )+
[0164] α2∑ j log(N k,j ) / (∑ j sgn(N k,j ))
[0165] Where α1, α2∈[0,1], and α1+α2=1.
[0166] Example 7
[0167] According to one embodiment of the present invention, such as Figure 6 As shown, an information processing device 600 is proposed, comprising: an acquisition unit 602 for acquiring text data and obtaining a target Uniform Resource Locator (URL) based on the text data; a first determination unit 604 for inputting the text data into a target business classification model to obtain the target business relevance of the text data; a second determination unit 606 for determining the popularity value of the target URL; a third determination unit 608 for determining the recommendation index of the target URL based on the popularity value and the target business relevance of the text data; and an output unit 610 for outputting the target URL based on the recommendation index.
[0168] The technical solution of this application proposes an information processing device 600, including an electronic device of the information processing device 600, which can label a recommendation index for target Uniform Resource Locators extracted from text data, and output the target Uniform Resource Locators according to the recommendation index, thereby realizing the diversified utilization of target Uniform Resource Locators.
[0169] Specifically, text data is acquired, and a target Uniform Resource Locator (URL) is obtained from the text data. During this process, the data contained in the text data is organized to identify sites, social media accounts, etc. that frequently publish information relevant to the target business.
[0170] In one technical solution, the text data can be an article sent by a user that is related to the target business. The article can include information such as leading figures in the industry or field in which the target business is located, research progress, and advertisements.
[0171] In one technical solution, text data is input into a classification model of the target business to obtain the relevance of the text data to the target business. In other words, the classification model of the target business is used to predict the category of the text data in order to determine the probability that the text data belongs to the target business.
[0172] In this technical solution, the classification model for the target business can be a pre-trained classification model. Specifically, it is a model trained using information that the target business is interested in and information that is not related to the target business.
[0173] In one of the technical solutions, the classification model for the target business is a binary classification model.
[0174] One technical solution includes cleaning the text data before inputting it into the target business classification model, such as removing target Uniform Resource Locators (URLs) and symbols like @ from the text data, in order to ensure the credibility of the target business relevance of the output text data.
[0175] In one technical solution, the popularity value of the target URI is determined. Specifically, this reflects the degree of diffusion of the target URI on different platforms and its occurrence on any platform. By statistically analyzing the popularity value of the target URI, recommendations can be made based on the popularity of the target URI when determining the recommendation index. This ensures that the recommended information sources are commonly used or frequently appearing sources, thus ensuring the credibility of the recommended information sources.
[0176] Furthermore, since the recommendation index is determined based on the popularity value and the relevance of the text data to the target business, it can ensure that the recommended information source has a strong correlation with the target business, thus achieving accurate information source recommendation.
[0177] In one of the technical solutions, the acquisition unit 602 is specifically used to: extract Uniform Resource Locators (URLs) from text data to obtain extraction results; determine the site type corresponding to the extraction results; and parse the extraction results based on the parsing rules corresponding to the site type to obtain the target URL.
[0178] In this technical solution, by obtaining the site type corresponding to the extraction result, the extraction result can be parsed according to the parsing rules corresponding to the site type, which ensures the accuracy of the obtained target Uniform Resource Locator and reduces the occurrence of situations where the extraction result is not parsed correctly.
[0179] In one of the technical solutions, when the site type is a social media site, the target Uniform Resource Locator includes the hostname domain and the path domain; when the site type is a non-social media site, the target Uniform Resource Locator includes the hostname domain.
[0180] In this technical solution, site types are divided into social media sites and non-social media sites. Under social media sites, the target uniform resource locator includes a hostname domain and a path domain. Since the target uniform resource locator also includes a path domain, it is possible to follow social media accounts, thus ensuring the credibility of the recommended information source.
[0181] Specifically, a Uniform Resource Locator typically consists of the following components in order: protocol field, hostname field, path field, filename field, and parameter field.
[0182] To ensure accurate parsing, the extracted results are categorized based on whether they belong to social media sites. If they belong to social media sites, the extracted results are parsed according to the parsing rules corresponding to social media sites to obtain the target Uniform Resource Locator (URL). If they do not belong to social media sites, the extracted results are parsed according to the parsing rules corresponding to non-social media sites to obtain the target URL.
[0183] Specifically, social media sites retain hostname and path domains. For example, the social media account https: / / twitter.com / rxxxa / status / 1xxx7?s=21 resolves to https: / / twitter.com / rxxxa, where rxxxxa represents the username, which is the part corresponding to the path domain.
[0184] Specifically, social media sites, which are essentially regular sites, retain the hostname domain. For example, https: / / www.kxxxr.com / xxx / mxxxy / lexxxa is re-resolved to https: / / www.kxxxr.com / .
[0185] One technical solution uses regular expressions to extract all Uniform Resource Locators (URLs) from the text data, thus obtaining the extraction results.
[0186] In one technical solution, a URL parser is used to parse the extraction results. The input of the URL parser is the extraction results, and the output is the website homepage, the homepage of social media people, organizations, etc.
[0187] In one of the technical solutions, the acquisition unit 602 is also used to: determine whether the extraction result is a short link; if so, to simulate a request to the extraction result in order to obtain the target Uniform Resource Locator.
[0188] In this technical solution, by determining whether the extracted result is a short link, a simulated request is made to the extracted result that is a short link if the determination result is yes, thereby obtaining the target Uniform Resource Locator (URL). This process expands the applicable scenarios for determining the target URL.
[0189] Specifically, given the word limits on content posted on social media platforms, longer links are usually shortened to a shorter link. Therefore, by determining whether the extracted result is a short link, the true Uniform Resource Locator can be displayed.
[0190] For example, the extracted short link https: / / t.co / 1cMxxxb?amp=1 is redirected to https: / / www.kxxxr.com / xxx / mxxxy / lexxxa.
[0191] In one of the technical solutions, the second determining unit 606 is further configured to: compare the target Uniform Resource Locator (URI) with a preset source database; determine whether the target URI does not exist in the preset source database; if so, store the target URI in the preset source database.
[0192] In this technical solution, the target Uniform Resource Locator (URI) is compared with a preset source database, and the preset source database is updated based on the comparison results. In this process, the discovery and updating of the target URI are realized, the preset source database is automatically maintained, and user involvement is reduced.
[0193] In one of the technical solutions, the third determining unit 608 is further used to: determine whether a target Uniform Resource Locator exists in the preset source database; if so, update the recommendation index of the target Uniform Resource Locator.
[0194] In this technical solution, when a target Uniform Resource Locator (URL) is detected in the preset source database, the recommendation index of the target URL is updated to ensure that the recommendation index is up-to-date. This ensures that when recommending sources based on the recommendation index, sources that are suitable for the current situation can be recommended, thus ensuring the credibility of the recommended sources.
[0195] In one of the technical solutions, the third determining unit 608 is specifically used to: obtain a first coefficient value and a second coefficient value; calculate the product of the first coefficient value and the popularity value to obtain a first calculated value; calculate the product of the average value of the target business relevance and the second coefficient value to obtain a second calculated value; and use the sum of the first calculated value and the second calculated value as the recommendation index of the target unified resource locator, wherein the sum of the first coefficient value and the second coefficient value is 1.
[0196] In this technical solution, by introducing a first coefficient value and a second coefficient value, the sum of which is 1, the average value of the popularity value and the relevance of the target business can be adjusted when calculating the recommendation index of the target Uniform Resource Locator. This ensures that the calculated recommendation index accurately reflects the recommendability of the target Uniform Resource Locator and guarantees the credibility of the recommendation.
[0197] Specifically, rect k =β1hotness k +β2avg(p k，j ).
[0198] Where, rect k This represents the recommendation index of a target Uniform Resource Locator (URL) with a unique index k, where β1 is the first coefficient value, β2 is the second coefficient value, and avg(p) represents the recommendation index. k，j () represents the average relevance of the target business.
[0199] Where β1, β2∈[0,1], and β1+β2=1.
[0200] In one of the technical solutions, the output unit 610 is specifically used to: determine the ranking of the recommendation index of the target Uniform Resource Locator in the recommendation set; and output the target Uniform Resource Locator according to the ranking.
[0201] In this technical solution, the recommendation indices in the recommendation set are sorted so that the output can be based on the sorting results. This ensures that the recommendation index of the target Uniform Resource Locator is relatively high, thus ensuring the credibility of the recommendation source.
[0202] In one of the technical solutions, the recommendation indices of the recommendation set are sorted from high to low.
[0203] In one of the technical solutions, the second determining unit 606 is further configured to: obtain a recommendation set in a preset source database; determine whether the target Uniform Resource Locator belongs to the recommendation set; and if so, calculate the popularity value of the target Uniform Resource Locator.
[0204] In this technical solution, by first determining whether the target Uniform Resource Locator (URL) belongs to the recommendation set, and then determining whether to calculate the popularity value of the target URL based on the determination result, the calculation of the popularity value of the target URL is avoided when the target URL does not belong to the recommendation set, thereby reducing the total number of times the popularity value is calculated.
[0205] Example 8
[0206] According to one embodiment of the present invention, such as Figure 7As shown, an electronic device 700 is proposed, including a processor 702, a memory 704, and a program or instructions stored in the memory 704 and executable on the processor 702. When the program or instructions are executed by the processor 702, they implement the steps of any of the above-mentioned information processing methods.
[0207] Specifically, the above information processing uses rule matching templates to extract Uniform Resource Locators (URLs). The rule matching templates can be implemented based on regular expressions. The rule matching templates are stored in rule base 1, and the parsing rules corresponding to the site types in the above information processing are stored in rule base 2.
[0208] like Figure 8 As shown, the operation of electronic device 700 includes:
[0209] Step 802: Customize the rule matching template, extract the Uniform Resource Locators (URLs) contained in the text data, and redirect the URLs to handle short links.
[0210] Step 804: Analyze social media personalities, organization homepages, and other website homepages;
[0211] Step 806: Update the preset source database;
[0212] Step 808: Recommend target Uniform Resource Locators (URLs) based on the target business relevance of the text data and the popularity and business relevance of the target URLs.
[0213] In step 802, a rule matching template is retrieved from rule base 1; in step 804, the parsing rules corresponding to the site type are retrieved from rule base 2; in step 808, a classification model for the target business is used for processing. Specifically, a binary classification model is constructed using text content from specific fields and business-irrelevant text content accumulated daily. The category of the text data is determined using the binary classification model, and the probability of belonging to the category of interest is used as the business relevance. The popularity value of target URIs already in the recommendation set is updated, and target URIs not in the recommendation set are added to the recommendation set, and the popularity value of the target URI is calculated; the business relevance of the target URI is updated or calculated; for each target URI in the recommendation set, the recommendation index is calculated by combining the popularity value of the target URI with the business relevance, so as to realize the recommendation of information sources.
[0214] In this embodiment, the needs for information source acquisition from social media accounts and traditional websites are considered, and the popularity of information sources and business relevance are combined to achieve filtering and recommendation of Uniform Resource Locators, thereby improving the effectiveness of information source recommendation.
[0215] In one embodiment, a readable storage medium is provided, on which a program or instructions are stored, which, when executed by a processor, implement the steps of the information processing method as described above.
[0216] In this embodiment, when the program or instructions on the readable storage medium are executed, the steps of any of the above-described information processing methods are implemented. Therefore, the readable storage medium has all the beneficial effects of any of the above-described information processing methods, which will not be elaborated here.
[0217] In the description of this invention, the term "a plurality of" refers to two or more. Unless otherwise explicitly defined, the terms "upper," "lower," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. The terms "connection," "installation," "fixing," etc., should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection or an indirect connection through an intermediate medium. For those skilled in the art, the specific meaning of the above terms in this invention can be understood according to the specific circumstances.
[0218] In the description of this invention, the terms "one embodiment," "some embodiments," "specific embodiment," etc., refer to a specific feature, structure, material, or characteristic described in connection with that embodiment or example, which is included in at least one embodiment or example of the invention. In this invention, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.
[0219] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. An information processing method, characterized in that, include: Acquire text data, and obtain the target Uniform Resource Locator (URL) based on the text data; The text data is input into the classification model of the target business to obtain the target business relevance of the text data, where the target business relevance is a value that represents the degree of relevance between the text data and the target business. Determine the popularity value of the target Uniform Resource Locator; Based on the popularity value and the relevance to the target business, determine the recommendation index of the target Uniform Resource Locator; Based on the recommendation index, output the target Uniform Resource Locator; The process of acquiring text data and obtaining a target Uniform Resource Locator (URL) based on the text data specifically includes: The text data is subjected to Uniform Resource Locator (URL) extraction to obtain the extraction result; Determine the site type corresponding to the extraction result; The extraction results are parsed based on the parsing rules corresponding to the site type to obtain the target Uniform Resource Locator; Based on the fact that the site type is a social media site, the target Uniform Resource Locator includes a hostname domain and a path domain; Since the site type is a non-social media site, the target Uniform Resource Locator includes a hostname domain.
2. The information processing method according to claim 1, characterized in that, The step of acquiring text data and obtaining a target Uniform Resource Locator (URL) based on the text data further includes: Based on the fact that the extraction result is a short link, a simulated request is made to the extraction result to obtain the target Uniform Resource Locator.
3. The information processing method according to claim 1, characterized in that, Before determining the popularity value of the target Uniform Resource Locator, the method further includes: The target Uniform Resource Locator (URL) is compared with a preset source database; Determine whether the target Uniform Resource Locator (URL) does not exist in the preset source database. If so, store the target URL in the preset source database.
4. The information processing method according to claim 3, characterized in that, Also includes: Based on the existence of the target Uniform Resource Locator in the preset source database, the recommendation index of the target Uniform Resource Locator is updated.
5. The information processing method according to claim 1, characterized in that, The step of determining the recommendation index of the target Uniform Resource Locator (URL) based on the popularity value and the target business relevance of the text data specifically includes: Obtain the first and second coefficient values; Calculate the product of the first coefficient value and the heat value to obtain the first calculated value; The second calculated value is obtained by multiplying the average value of the target business relevance with the second coefficient value; The sum of the first calculated value and the second calculated value is used as the recommendation index for the target Uniform Resource Locator. The sum of the first coefficient value and the second coefficient value is 1.
6. The information processing method according to claim 1, characterized in that, The step of outputting the target Uniform Resource Locator (URL) based on the recommendation index specifically includes: Determine the ranking of the recommendation index of the target Uniform Resource Locator in the recommendation set; Based on the sorting, output the target Uniform Resource Locator.
7. The information processing method according to claim 6, characterized in that, Before determining the popularity value of the target Uniform Resource Locator, the method further includes: Retrieve the recommended set from the preset source database; Determine whether the target Uniform Resource Locator belongs to the recommendation set. If it does, calculate the popularity value of the target Uniform Resource Locator.
8. An information processing device, characterized in that, include: The acquisition unit is used to acquire text data and obtain a target Uniform Resource Locator (URL) based on the text data. The first determining unit is used to input the text data into the classification model of the target business to obtain the target business relevance of the text data; The second determining unit is used to determine the heat value of the target Uniform Resource Locator; The third determining unit is used to determine the recommendation index of the target Uniform Resource Locator based on the popularity value and the target business relevance of the text data; The output unit is used to output the target Uniform Resource Locator (URL) based on the recommendation index. The process of acquiring text data and obtaining a target Uniform Resource Locator (URL) based on the text data specifically includes: The text data is subjected to Uniform Resource Locator (URL) extraction to obtain the extraction result; Determine the site type corresponding to the extraction result; The extraction results are parsed based on the parsing rules corresponding to the site type to obtain the target Uniform Resource Locator; Based on the fact that the site type is a social media site, the target Uniform Resource Locator includes a hostname domain and a path domain; Since the site type is a non-social media site, the target Uniform Resource Locator includes a hostname domain.
9. An electronic device, characterized in that, It includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the information processing method as described in any one of claims 1 to 7.
10. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the information processing method as described in any one of claims 1 to 7.