Method, device and equipment for determining content quality and storage medium

Abstract

The disclosure provides a content quality determination method, device and equipment and a storage medium, relates to the computer field, and in particular to the fields of big data, the Internet and the like. The content quality determination method comprises: obtaining source identification information of a first target content; determining a production source of the first target content according to the source identification information; and determining content quality of the production source according to quality evaluation characteristics of the production source. According to the content quality determination method of the disclosure, the production source of the first target content can be mined according to the source identification information of the first target content, and the content quality of the production source can be determined according to the quality evaluation characteristics of the production source, so that the efficiency is higher, and the discrimination cost can be reduced.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, in particular to technical fields such as big data and the Internet, and specifically to a method for determining content quality, an apparatus for determining content quality, an electronic device, a computer-readable storage medium, and a computer program product. Background Art

[0002] In related technologies, with the continuous development of Internet technology, there is a massive amount of content in the Internet field, including a large amount of automatically generated low-quality content. Currently, the main method for determining content quality is to conduct single-page judgment, which is inefficient and costly. Summary of the Invention

[0003] The present disclosure provides a method, apparatus, device, and storage medium for determining content quality.

[0004] According to a first aspect of the present disclosure, a method for determining content quality is provided, comprising:

[0005] Obtaining source identification information of the first target content;

[0006] determining a production source of the first target content according to the source identification information;

[0007] Determine the content quality of the production source based on the quality evaluation characteristics of the production source.

[0008] According to a second aspect of the present disclosure, there is provided a device for determining content quality, comprising:

[0009] An acquisition module, configured to acquire source identification information of the first target content;

[0010] A first determining module, configured to determine a production source of the first target content according to the source identification information;

[0011] The second determination module is used to determine the content quality of the production source according to the quality evaluation characteristics of the production source.

[0012] According to a third aspect of the present disclosure, there is provided an electronic device, including:

[0013] a memory communicatively connected to at least one processor; wherein,

[0014] The memory stores instructions that can be executed by at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to perform the method of any embodiment of the above aspects.

[0015] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, where the computer instructions are used to enable a computer to execute the method of any one of the above-mentioned embodiments.

[0016] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program, which implements the method of any one of the above-mentioned embodiments when executed by a processor.

[0017] According to the technical solution disclosed in the present invention, the production source of the first target content can be mined based on the source identification information of the first target content, and the content quality of the production source can be determined based on the quality evaluation characteristics of the production source, which is more efficient and can reduce the judgment cost.

[0018] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are provided to facilitate a better understanding of the present invention and do not constitute a limitation of the present disclosure.

[0020] Figure 1 is a flowchart of a method for determining content quality according to an embodiment of the present disclosure;

[0021] Figure 2 is a schematic diagram of a process for determining the content quality of a production source according to the first embodiment of the present disclosure;

[0022] Figure 3 is a schematic diagram of a process for determining the content quality of a production source according to a second embodiment of the present disclosure;

[0023] Figure 4 is a schematic diagram of a process for determining the content quality of a production source according to a third embodiment of the present disclosure;

[0024] Figure 5 is a schematic diagram of a process for determining the content quality of a production source according to a fourth embodiment of the present disclosure;

[0025] Figure 6 is a schematic diagram of a process for determining the content quality of a production source according to a fifth embodiment of the present disclosure;

[0026] Figure 7 is a schematic diagram of a process for determining the content quality of a production source according to a sixth embodiment of the present disclosure;

[0027] Figure 8 is a schematic diagram of a process for determining the content quality of a production source according to a seventh embodiment of the present disclosure;

[0028] Figure 9 is an example diagram of an application of a method for determining content quality according to an embodiment of the present disclosure;

[0029] Figure 10 is a structural block diagram of a device for determining content quality according to an embodiment of the present disclosure;

[0030] Figure 11 The present invention is a block diagram of an electronic device for implementing the method for determining content quality according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0031] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0032] like Figure 1 As shown, the method for determining content quality according to the embodiment of the first aspect of the present disclosure includes:

[0033] Step S101: Obtain source identification information of the first target content;

[0034] Step S102: determining the production source of the first target content according to the source identification information;

[0035] Step S103: Determine the content quality of the production source based on the quality evaluation characteristics of the production source.

[0036] For example, the first target content may be obtained from a web page, i.e., the first target content may be page content, such as text, images, or icons, etc. Alternatively, the first target content may be obtained from a website, i.e., the first target content may be website content, such as text, images, or icons, etc. Of course, obtaining the first target content is not limited to this.

[0037] The source identification information may be a Uniform Resource Locator (URL), account registration information, entity filing information, etc.

[0038] Based on the source identification information, the production source of the first target content, that is, the source of the first target content, can be determined. The production source can be a site, a primary domain, an account, or an entity.

[0039] The quality evaluation features may be preset, such as at least one of page growth rate, publishing time distribution rules, historical traffic, content domain, page layout type, and user behavior characteristics. Based on the at least one quality evaluation feature, the quality of the content of the production source may be determined.

[0040] The content quality of a production source can include low quality and high quality. Low quality can include junk cheating content and low-value content. Based on the content quality of a production source, the production source can be determined as a junk cheating source, a low-value source, or a high-value source. A junk cheating source can be a source that automatically produces content. Junk cheating content has characteristics such as confusing article logic, poor grammar, and inconsistent titles and content. The content quality of a low-value source is higher than that of a junk cheating source, but its content is not high-quality content. For example, user A asks the question "What is the normal body temperature of a human body" on Baidu Tieba, and user B answers "36.5℃." User B's answer corresponds to user A's question, and its content quality is higher than that of junk cheating content, but it is not high-quality content. The content quality of a high-value source is higher than that of a low-value source, and its content is artificially produced high-quality content.

[0041] In one example, the production source can be a website. In this case, the source identification information of the first target content can be a URL. Since every information resource has a unified and unique address online, this address is the URL, or network address. Based on the URL of the first target content, the website that produced the first target content can be determined. Based on the quality evaluation characteristics of the website, the website can be determined to be a spam source, a low-value source, or a high-value source.

[0042] In one example, the production source can be a primary domain. In this case, the source identification information of the first target content can also be a URL. Based on the URL of the first target content, the primary domain that produced the first target content can be determined. Based on the quality evaluation characteristics of the primary domain, the primary domain can be determined to be a spam source, a low-value source, or a high-value source.

[0043] In one example, the production source may be an account. The account may be a personal account that directly publishes information on various platforms, such as Penguin Account, Baijia Account, etc. Alternatively, the account may also be an account that pushes data on various platforms (such as an inclusion platform). In this case, the account may be associated with at least one site or main domain. The source identification information of the first target content may be the registration information of the account. Based on the registration information of the account, the account that produced the first target content may be determined. Based on the quality evaluation characteristics of the account, the account may be determined to be a spam cheating source, a low-value source, or a high-value source.

[0044] In one example, the production source can also be an entity. For example, the entity can be an enterprise or institution registered with the Ministry of Industry and Information Technology. In this case, the source identification information of the first target content can be the entity's registration information. Based on the entity's registration information, the entity that produced the first target content can be determined. Based on the entity's quality evaluation characteristics, the entity can be determined to be a spam source, a low-value source, or a high-value source.

[0045] After determining the quality of the production source's content, the source's content can be processed accordingly. For example, if the production source is a spam or cheating source, the already collected content can be cleaned up and suppressed, and the source can no longer be included. If some of the content in the production source is low-quality and some is high-quality, a flexible suppression can be performed, that is, the already collected low-quality content is cleaned up and suppressed, the high-quality content is retained, and the inclusion quota of the production source is reduced.

[0046] According to the content quality determination method of the embodiment of the present disclosure, the production source of the first target content can be mined based on the source identification information of the first target content, and the content quality of the production source can be determined based on the quality evaluation characteristics of the production source. Compared with the content quality determination method for a single page, the content quality can be determined from the source, with a wider and more effective processing range, higher efficiency, and can reduce the cost of determination. Furthermore, low-quality production sources can be directly suppressed accordingly, thereby effectively reducing low-quality content on the Internet.

[0047] In one embodiment, Figure 2 As shown, the quality evaluation feature includes the page growth rate. In step S103, the content quality of the production source is determined based on the quality evaluation feature of the production source, which may include:

[0048] Step S201: determining a page growth rate based on the publishing time of multiple pages of a production source;

[0049] Step S202: comparing the page growth rate with a preset growth rate;

[0050] Step S203: Determine the content quality of the production source based on the comparison result.

[0051] For example, based on the publication time of multiple pages of a production source, the daily page growth rate of the production source can be determined, and the daily page growth rate can be compared with a preset daily page growth rate, where the preset daily page growth rate can be pre-set based on the normal manual production speed. If the daily page growth rate of the production source is much greater than the preset daily page growth rate, it indicates that the production source's productivity, push volume, and publication volume, and other behavioral data are abnormal. It can be determined that the page content of the production source is likely low-quality content, and the production source is likely a spam or cheating source. In this case, the included content can be cleaned up and suppressed, and the production source can no longer be included.

[0052] For example, if the production source is a personal account, the preset daily page growth rate can be 1 to 80 pages. If the daily page growth rate is 1 to 80, it indicates that the personal account's page production rate is normal, and it can be determined that the personal account is likely a high-value source. If the daily page growth rate is hundreds, thousands, or even tens of thousands, the account's daily page growth rate is much greater than the preset daily page growth rate. In this case, it can be determined that the personal account's productivity and publishing volume are abnormal, and its production capacity far exceeds normal human production capacity, thus determining that the personal account is likely a spam source.

[0053] If the production source is a larger site, it has more channels and sources, and the preset daily page growth rate can be 100,000 to 2 million pages. If the number of pages increases by 100,000 to 2 million per day, it means that the site's page production speed is normal manual production speed, and it can be determined that the site is likely a high-value source. If the number of pages increases by tens of millions or even hundreds of millions per day, the daily page growth rate of the site is far greater than the preset daily page growth rate. In this case, it can be determined that the site's productivity is abnormal. Many pages on the site are likely mass-produced and worthless pages, and therefore the site is likely a spam source.

[0054] Therefore, through the above settings, the content quality of the entire production source can be directly determined by comparing the page growth rate with the preset growth rate. The judgment method is simple, and the efficiency of determining content quality can be improved, effectively reducing the cost of determining content quality.

[0055] In one embodiment, reference Figure 3 The quality evaluation characteristics include a release time distribution rule. In step S103, determining the content quality of the production source based on the quality evaluation characteristics of the production source may include:

[0056] Step S301: determining a publishing time distribution rule based on the publishing time of multiple pages of a production source;

[0057] Step S302: Compare the release time distribution rule with the preset distribution rule;

[0058] Step S303: Determine the content quality of the production source based on the comparison result.

[0059] Exemplarily, when the publishing time distribution rules of multiple pages of a production source meet the preset distribution rules, it can be determined that the production source is more likely to be a high-value source; when the publishing time distribution rules of multiple pages of a production source do not meet the preset distribution rules, it can be determined that the production source is more likely to be a spam cheating source.

[0060] Among them, the preset distribution rules can be pre-set according to normal working hours. For example, when the publishing time of multiple pages of the production source is basically concentrated in working hours (for example, 8:00-18:00), it means that the multiple pages of the production source are more likely to be manually produced pages, and it can be determined that the site is more likely to be a high-value source. When the publishing time distribution of multiple pages of the production source is relatively scattered and not concentrated in working hours, it can be determined that the publishing time distribution of the production source is abnormal, which means that the production source is more likely to automatically produce content, and the production source is more likely to be a spam cheating source.

[0061] The method of the above-mentioned embodiment of the present disclosure can directly determine the content quality of the entire production source by comparing the publishing time distribution rules of multiple pages with the preset distribution rules, without judging the content of each page. The judgment method is simple, and it can also improve the efficiency of determining the content quality and reduce the cost of determining the content quality.

[0062] In one embodiment, combined Figure 4 The quality evaluation feature includes historical traffic. In step S103, determining the content quality of the production source based on the quality evaluation feature of the production source may include:

[0063] Step S401: Obtain a historical flow curve of a production source within at least one specified time period;

[0064] Step S402: Determine the content quality of the production source based on the historical traffic curve.

[0065] It should be noted that the above-mentioned "traffic" can refer to the number of visits to the production source (such as a site), which can be understood as an indicator used to describe the number of users visiting the production source and the number of pages browsed by users. Commonly used statistical indicators include the number of independent users of the production source, the total number of users (including repeat visitors), the number of web page views, the number of page views per user, the average time users stay on the website, etc.

[0066] For example, since the spam cheating source of automatically produced content may be a long-term idle site or a purchased site, the historical traffic of the production source is usually low or has large fluctuations. After obtaining the historical traffic curve of the production source in a specified time period, if the historical traffic of the production source is relatively high and stable, it is determined that the production source is likely to be a high-value source. If the historical traffic of the production source is low, or the traffic of the production source suddenly increases significantly before holidays, or the traffic of the production source suddenly decreases significantly during major events, it means that the production source is likely to be a spam cheating source.

[0067] For example, within a specified time period, if the number of independent users of a production source remains basically stable at 3,000 to 5,000, it is determined that the production source is most likely a high-value source; if the number of independent users of a production source is basically 0 to 10, it means that the production source is likely to be idle for a long time, and it is determined that the production source is most likely a spam cheating source; if the number of independent users of a production source is usually 0 to 100, and suddenly increases to more than 5,000 before the Spring Festival, it means that the production source may be trying to obtain higher profits before the Spring Festival. At this time, it can be determined that the production source is most likely a spam cheating source; if the number of independent users of a production source is usually 200 to 500, and the traffic suddenly drops to below 20 during major events, due to stricter supervision during major events, it can be determined that the production source is most likely a spam cheating source that automatically produces content.

[0068] In this way, the content quality of the production source can be directly determined based on the historical traffic curve. The historical traffic curve is more intuitive and easier to obtain. There is no need to judge each page in the production source, which can further improve the efficiency of determining the content quality of the production source and reduce the cost of determining the content quality of the production source.

[0069] In one embodiment, Figure 5 As shown, the quality evaluation feature includes the domain to which the content belongs. In step S103, the content quality of the production source is determined based on the quality evaluation feature of the production source, which may include:

[0070] Step S501: determining the domain to which the contents contained in the multiple pages of the production source belong;

[0071] Step S502: Compare the number of categories of the content field with the number of preset categories;

[0072] Step S503: Determine the content quality of the production source based on the comparison result.

[0073] Exemplarily, when the number of categories of the content fields contained in the multiple pages of the production source is less than or equal to the preset number of categories, it can be determined that the production source is more likely to be a high-value source; when the number of categories of the content fields contained in the multiple pages of the production source is much greater than the preset number of categories, it can be determined that the production source is more likely to be a spam cheating source that automatically produces content.

[0074] For example, if the production source is a website, the number of preset categories can be 1 to 3. When the website includes 100 pages, it is necessary to determine the content fields of each of the 100 pages. If the content fields of all 100 pages are sports, the number of categories of the content fields contained in the 100 pages is 1, indicating that the website is focused on the sports field, and it can be determined that the production source is likely a high-value source. If the content fields corresponding to the 100 pages include 30 fields such as sports, entertainment, and military, it means that the distribution of the content fields contained in the 100 pages of the website fluctuates greatly and is scattered, and it can be determined that the production source is likely a spam source.

[0075] Therefore, the content quality of the production source can be determined by comparing the fields to which the content contained in multiple pages belongs with the number of preset categories, so that the content quality of the production source can be judged from the dimensions of multiple pages. Compared with the single-page judgment method, the judgment efficiency can be effectively improved. In the case that the production source is a spam cheating source, all pages of the entire production source can be suppressed in batches, and the processing range is wider, which can effectively reduce the low-quality spam content on the Internet.

[0076] In one embodiment, reference Figure 6 The quality evaluation feature includes the page layout type. In step S103, the content quality of the production source is determined based on the quality evaluation feature of the production source, which may include:

[0077] Step S601: determining the page layout types corresponding to the multiple pages of the production source;

[0078] Step S602: Determine the content quality of the production source according to the typesetting type of each page.

[0079] In one example, the page layout type may include page style (including background image, font style, color, etc. of title and body). For example, if the page styles of multiple pages are basically the same, it indicates that the multiple pages of the production source are professionally designed, and it can be determined that the production source is likely a high-value source; if the page styles of multiple pages are relatively chaotic, it indicates that the multiple pages of the production source are not professionally designed, and it can be determined that the production source is likely a spam source.

[0080] In another example, the page layout type can include templates. In the case of a spam cheating source, the multiple pages within the production source are typically obtained by crawling internet webpages and splicing and combining the content of multiple webpages. In this case, the multiple pages within the production source may be directly adjusted based on the source code, resulting in different and chaotic template styles for the multiple pages. Therefore, if the templates corresponding to the multiple pages of the production source are chaotic and diverse, it can be determined that the production source is likely a spam cheating source.

[0081] With this setting, the content quality of the production source can be determined based on the page layout types of multiple pages of the production source, so that the content quality of the production source can also be judged from the dimensions of multiple pages, which can improve the judgment efficiency and then perform large-scale suppression processing on all pages of low-quality production sources, effectively reducing spam content on the Internet.

[0082] In one embodiment, combined Figure 7 The quality evaluation characteristics include user behavior characteristics, which include user click volume and / or user reading time. In step S103, determining the content quality of the production source based on the quality evaluation characteristics of the production source may include:

[0083] Step S701: determining user behavior characteristics based on the user click volume and / or user reading time of the production source;

[0084] Step S702: Determine the content quality of the production source based on user behavior characteristics.

[0085] For example, if the user clicks on each page of the production source are high and / or the user reading time is long, the user behavior performance is good, and the production source is more likely to be a high-value source. If the user clicks on each page of the production source are low and / or the user reading time is short, the user behavior performance is poor, and the production source may be a spam source that automatically produces content.

[0086] Therefore, through the above settings, the content quality of the production source can be judged from the dimension of the user's behavioral characteristics, which can increase the ways to determine the content quality of the production source, thereby effectively determining the junk cheating sources that produce low-quality content and reducing the cost of determining the junk cheating sources.

[0087] In one embodiment, Figure 9 As shown, in step S103, determining the content quality of the production source according to the quality evaluation characteristics of the production source may include: inputting at least one quality evaluation characteristic of the production source into a content quality evaluation model to determine the content quality of the production source.

[0088] Exemplarily, the model training needs positive samples and negative samples. When obtaining the positive samples, a page with high user click volume and long user reading time can be selected as a positive sample, or the top N pages of a search engine can be directly captured as positive samples. Wherein, N can be determined according to actual needs, for example, N can be 3-5.

[0089] The cost of obtaining negative samples is high, so negative samples can be automatically produced according to the way of automatically producing low-quality content by the garbage cheating source. For example, a plurality of pages can be collected, and the sentences and / or paragraphs of the content of each page can be randomly shuffled, spliced, deleted, etc., and then the texts of the plurality of pages are spliced and combined to obtain a plurality of new pages, so as to take the plurality of new pages as negative samples.

[0090] The content quality evaluation model can output the final result, for example, it can output the probability of corresponding production source being a garbage cheating source, a low-value source and a high-value source, so as to determine the content quality of the corresponding production source.

[0091] Therefore, by inputting the quality evaluation features of the production source into the content quality evaluation model, the content quality of the production source can be more accurately determined, avoiding misjudgment, and effectively improving the accuracy and reliability of determining the content quality of the production source.

[0092] In an embodiment, as shown in Figure 8 In step S103, according to the quality evaluation features of the production source, the content quality of the production source can be determined, which can include:

[0093] Step S801: determining a plurality of candidate pages from a plurality of pages of the production source;

[0094] Step S802: determining whether the candidate page is a target page according to the quality evaluation features of the candidate page, wherein the target page includes second target content that does not meet the preset evaluation standard;

[0095] Step S803: determining the content quality of the production source according to the proportion of the target page in the plurality of pages.

[0096] For example, when a production source includes 100 pages, the 100 pages can be divided into 10 candidate groups, each of which includes 10 candidate pages. First, one candidate group can be determined from the 100 pages. Then, based on the page evaluation characteristics of each candidate page in the candidate group, it is determined whether the 10 candidate pages in the candidate group are target pages. This process is repeated until all candidate pages in the 10 candidate groups have been tested. Finally, the ratio of the number of target pages to the total number of pages is calculated. For example, if the 10 candidate groups include 90 target pages, then the ratio of the number of target pages to the total number of pages is 90%, indicating that the production source is likely to be a spam or cheating source and can be suppressed.

[0097] Therefore, the above-mentioned embodiment of the present disclosure can determine the proportion of low-quality content in the production source according to the proportion of the target page in multiple pages, so as to more accurately judge the content quality of the production source.

[0098] In one embodiment, the above-mentioned quality evaluation features include at least one of page growth rate, release time distribution rules, historical traffic, content domain, page layout type, and user behavior characteristics. In this way, when there is only one quality evaluation feature, the judgment method is simpler, which can improve the efficiency of determining the content quality of the production source; when there are multiple quality evaluation features, the content quality of the production source can be determined by multiple dimensions such as the production source's behavioral characteristics (such as page growth rate, release time distribution rules, historical traffic, etc.), multi-page characteristics (such as the content domain, page layout type, etc.), and user behavior characteristics, with higher accuracy.

[0099] In an optional embodiment, the quality evaluation features may also include single-page features, including paragraph logic and / or title-text consistency. In step S103, determining the content quality of the production source based on the quality evaluation features of the production source may include: determining the content quality of the production source based on the paragraph logic and / or title-text consistency of each page of the production source. Thus, the content quality of the production source can be determined from the features of a single page. In the case where each page is unreadable due to confusing logic or grammatical errors, and / or the title and content are inconsistent, it can be determined that the production source is likely to be a spam or cheating source, thereby improving the reliability of determining the content quality of the production source.

[0100] like Figure 10 As shown, the content quality determination device 1000 according to the second embodiment of the present disclosure includes:

[0101] An acquisition module 1001 is configured to acquire source identification information of a first target content;

[0102] A first determining module 1002 is configured to determine a production source of the first target content according to the source identification information;

[0103] The second determining module 1003 is configured to determine the content quality of the production source according to the quality evaluation characteristics of the production source.

[0104] In one embodiment, the quality evaluation feature includes page growth rate, and the second determination module 1003 includes:

[0105] A speed determination submodule is used to determine the page growth speed based on the publishing time of multiple pages of the production source;

[0106] The speed comparison submodule is used to compare the page growth speed with the preset growth speed;

[0107] The first quality determination submodule is configured to determine the content quality of the production source according to the comparison result.

[0108] In one embodiment, the quality evaluation feature includes a release time distribution rule, and the second determination module 1003 includes:

[0109] A rule determination submodule is used to determine a publishing time distribution rule based on the publishing time of multiple pages of the production source;

[0110] A rule comparison submodule is used to compare the release time distribution rule with the preset distribution rule;

[0111] The second quality determination submodule is configured to determine the content quality of the production source according to the comparison result.

[0112] In one embodiment, the quality evaluation feature includes historical traffic, and the second determination module 1003 includes:

[0113] A curve acquisition submodule is used to obtain a historical flow curve of a production source within at least one specified time period;

[0114] The third quality determination submodule is used to determine the content quality of the production source based on the historical traffic curve.

[0115] In one embodiment, the quality evaluation feature includes the domain to which the content belongs, and the second determination module 1003 includes:

[0116] A domain determination submodule is used to determine the domain to which the content contained in multiple pages of the production source belongs;

[0117] A quantity comparison submodule is used to compare the number of categories in the field to which the content belongs with the number of preset categories;

[0118] The fourth quality determination submodule is configured to determine the content quality of the production source according to the comparison result.

[0119] In an implementation, the quality evaluation feature includes a page layout type, and the second determination module 1003 includes:

[0120] a type determination sub-module, configured to determine a page layout type corresponding to each of the plurality of pages of the production source;

[0121] a fifth quality determination sub-module, configured to determine the content quality of the production source according to the page layout type.

[0122] In an implementation, the quality evaluation feature includes a user behavior feature, and the user behavior feature includes a user click volume and / or a user reading duration; the second determination module 1003 includes:

[0123] a behavior feature determination sub-module, configured to determine the user behavior feature according to the user click volume and / or the user reading duration of the production source;

[0124] a sixth quality determination sub-module, configured to determine the content quality of the production source according to the user behavior feature.

[0125] In an implementation, the second determination module 1003 is further configured to input the quality evaluation feature of the production source into a content quality evaluation model to determine the content quality of the production source.

[0126] In an implementation, the second determination module 1003 includes:

[0127] a candidate page determination sub-module, configured to determine a part of candidate pages from the plurality of pages of the production source;

[0128] a target page determination sub-module, configured to determine whether the candidate page is a target page according to the quality evaluation feature of the part of candidate pages, wherein the target page includes a second target content that does not meet a preset evaluation standard;

[0129] a seventh quality determination sub-module, configured to determine the content quality of the production source according to a proportion of the target page in the plurality of pages.

[0130] In an implementation, the quality evaluation feature includes at least one of a page growth speed, a publishing time distribution rule, a historical traffic, a content belonging field, a page layout type, and a user behavior feature.

[0131] In an implementation, the quality evaluation feature can further include a single-page feature, and the single-page feature includes a paragraph logic and / or a title-text consistency; the second determination module 1003 is further configured to determine the content quality of the production source according to the paragraph logic and / or the title-text consistency of each page of the production source.

[0132] The functions and effects of each module or sub-module in each device of the embodiments of the present disclosure can be found in the corresponding description in the above method embodiments, and will not be repeated here.

[0133] In the technical solutions disclosed herein, the acquisition, storage, and application of user personal information involved comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0134] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0135] Figure 11 A schematic block diagram of an example electronic device 1100 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are provided as examples only and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0136] like Figure 8 As shown, the device 1100 includes a computing unit 1101, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1102 or a computer program loaded from a storage unit 1108 into a random access memory (RAM) 1103. Various programs and data required for the operation of the device 1100 can also be stored in the RAM 1103. The computing unit 1101, the ROM 1102, and the RAM 1103 are connected to each other via a bus 1104. An input / output (I / O) interface 1105 is also connected to the bus 1104.

[0137] Various components in device 1100 are connected to I / O interface 1105, including an input unit 1106, such as a keyboard and mouse; an output unit 1107, such as various types of displays and speakers; a storage unit 1108, such as a magnetic disk and optical disk; and a communication unit 1109, such as a network card, a modem, a wireless communication transceiver, etc. Communication unit 1109 allows device 1100 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0138] The computing unit 1101 may be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the computing unit 1101 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 1101 performs the various methods and processes described above, such as the method for determining content quality. For example, in some embodiments, the method for determining content quality may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 1108. In some embodiments, part or all of the computer program may be loaded and / or installed onto the device 1100 via the ROM 1102 and / or the communication unit 1109. When the computer program is loaded into the RAM 1103 and executed by the computing unit 1101, one or more steps of the method for determining content quality described above may be performed. Alternatively, in other embodiments, the computing unit 1101 may be configured to execute the content quality determination method in any other appropriate manner (for example, by means of firmware).

[0139] Various implementations of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-a-chip systems (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0140] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0141] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0142] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0143] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0144] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0145] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not a limitation herein.

[0146] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A method for determining content quality, comprising: Obtaining source identification information of the first target content; determining a production source of the first target content according to the source identification information; Determining the content quality of the production source based on quality evaluation characteristics of the production source; wherein the quality evaluation characteristics include at least one of page growth rate, publishing time distribution rules, and historical traffic; The step of determining the content quality of the production source based on the quality evaluation characteristics of the production source includes: At least one quality evaluation feature of the production source is input into a content quality evaluation model to determine the content quality of the production source; wherein, the training of the model requires positive samples and negative samples, and the pages ranked in the top N positions of the search engine are captured as the positive samples; multiple pages are collected, and the content sentences and / or paragraphs of each page are randomly shuffled, spliced ​​and deleted, and then the texts of the multiple pages are spliced ​​and combined to obtain multiple new pages, and the multiple new pages are used as the negative samples.

2. The method according to claim 1, wherein The quality evaluation feature includes a page growth rate. Determining the content quality of the production source based on the quality evaluation feature of the production source includes: determining the page growth rate according to the publishing time of the plurality of pages of the production source; comparing the page growth rate with a preset growth rate; The content quality of the production source is determined based on the comparison results.

3. The method according to claim 1, wherein The quality evaluation characteristics include a release time distribution rule. Determining the content quality of the production source based on the quality evaluation characteristics of the production source includes: Determining the publishing time distribution rule according to the publishing time of the plurality of pages of the production source; comparing the release time distribution rule with a preset distribution rule; The content quality of the production source is determined based on the comparison results.

4. The method according to claim 2, wherein: The quality evaluation feature includes historical traffic. Determining the content quality of the production source based on the quality evaluation feature of the production source includes: Obtaining a historical flow curve of the production source within at least one specified time period; The content quality of the production source is determined according to the historical traffic curve.

5. The method according to claim 1, wherein The quality evaluation feature includes the domain to which the content belongs. Determining the content quality of the production source based on the quality evaluation feature of the production source includes: determining the domain to which the contents contained in the plurality of pages of the production source belong; Comparing the number of categories in the field to which the content belongs with the number of preset categories; The content quality of the production source is determined based on the comparison results.

6. The method according to claim 1, wherein The quality evaluation feature includes a page layout type. Determining the content quality of the production source based on the quality evaluation feature of the production source includes: Determining page layout types corresponding to the plurality of pages of the production source; The content quality of the production source is determined based on each of the page layout types.

7. The method according to claim 1, wherein The quality evaluation characteristics include user behavior characteristics; Determining the content quality of the production source based on the quality evaluation characteristics of the production source includes: Determining the user behavior characteristics based on the user click volume and / or user reading time of the production source; The content quality of the production source is determined according to the user behavior characteristics.

8. The method according to claim 1, wherein Determining the content quality of the production source based on the quality evaluation characteristics of the production source includes: Determining a portion of pages to be selected from the plurality of pages of the production source; determining, based on quality evaluation features of the plurality of pages to be selected, whether the page to be selected is a target page, wherein the target page includes second target content that does not meet a preset evaluation criterion; The content quality of the production source is determined according to the proportion of the target page in the multiple pages.

9. A device for determining content quality, comprising: An acquisition module, configured to acquire source identification information of the first target content; a first determining module, configured to determine a production source of the first target content according to the source identification information; A second determination module is configured to determine the content quality of the production source based on quality evaluation characteristics of the production source; wherein the quality evaluation characteristics include at least one of page growth rate, publishing time distribution rules, and historical traffic; Wherein, the second determining module is further configured to: At least one quality evaluation feature of the production source is input into a content quality evaluation model to determine the content quality of the production source; wherein, the training of the model requires positive samples and negative samples, and the pages ranked in the top N positions of the search engine are captured as the positive samples; multiple pages are collected, and the content sentences and / or paragraphs of each page are randomly shuffled, spliced ​​and deleted, and then the texts of the multiple pages are spliced ​​and combined to obtain multiple new pages, and the multiple new pages are used as the negative samples.

10. The device according to claim 9, wherein The quality evaluation feature includes page growth rate, and the second determination module includes: A speed determination submodule, configured to determine a page growth speed according to the publishing time of the plurality of pages of the production source; A speed comparison submodule, configured to compare the page growth speed with a preset growth speed; The first quality determination submodule is configured to determine the content quality of the production source according to the comparison result.

11. The device according to claim 9, wherein The quality evaluation feature includes a release time distribution rule, and the second determination module includes: A rule determination submodule, configured to determine the publishing time distribution rule according to the publishing time of the plurality of pages of the production source; A rule comparison submodule, configured to compare the release time distribution rule with a preset distribution rule; The second quality determination submodule is configured to determine the content quality of the production source according to the comparison result.

12. The device according to claim 9, wherein The quality evaluation feature includes historical traffic, and the second determination module includes: A curve acquisition submodule, configured to acquire a historical flow curve of the production source within at least one specified time period; The third quality determination submodule is configured to determine the content quality of the production source according to the historical traffic curve.

13. The device according to claim 9, wherein The quality evaluation feature includes the domain to which the content belongs, and the second determination module includes: A domain determination submodule, configured to determine the domain to which the contents contained in the plurality of pages of the production source belong; A quantity comparison submodule, used to compare the number of categories in the field to which the content belongs with the number of preset categories; The fourth quality determination submodule is configured to determine the content quality of the production source according to the comparison result.

14. The device according to claim 9, wherein The quality evaluation feature includes a page layout type, and the second determination module includes: A type determination submodule, configured to determine the page layout types corresponding to the plurality of pages of the production source; The fifth quality determination submodule is used to determine the content quality of the production source according to each of the page layout types.

15. The device according to claim 9, wherein The quality evaluation feature includes a user behavior feature, and the user behavior feature includes a user click volume and / or a user reading time. The second determination module includes: A behavior feature determination submodule, configured to determine the user behavior feature based on the user click volume and / or user reading time of the production source; The sixth quality determination submodule is configured to determine the content quality of the production source according to the user behavior characteristics.

16. The device according to claim 9, wherein The second determining module includes: A to-be-selected page determination submodule, configured to determine some to-be-selected pages from the plurality of pages in the production source; a target page determination submodule, configured to determine whether the to-be-selected page is a target page based on quality evaluation characteristics of the to-be-selected pages, wherein the target page includes second target content that does not meet a preset evaluation standard; The seventh quality determination submodule is configured to determine the content quality of the production source according to the proportion of the target page in the plurality of pages.

17. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 8.

18. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-8.

19. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 8.

Images