CN108717469B - Post sorting method, device and equipment and computer readable storage medium - Google Patents
Post sorting method, device and equipment and computer readable storage medium Download PDFInfo
- Publication number
- CN108717469B CN108717469B CN201810597468.1A CN201810597468A CN108717469B CN 108717469 B CN108717469 B CN 108717469B CN 201810597468 A CN201810597468 A CN 201810597468A CN 108717469 B CN108717469 B CN 108717469B
- Authority
- CN
- China
- Prior art keywords
- post
- sorting
- index
- ranking
- posts
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000012163 sequencing technique Methods 0.000 claims abstract description 12
- 230000006399 behavior Effects 0.000 claims description 70
- 238000012549 training Methods 0.000 claims description 48
- 238000012216 screening Methods 0.000 claims description 38
- 238000004891 communication Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000001174 ascending effect Effects 0.000 description 7
- 230000000694 effects Effects 0.000 description 5
- 238000010606 normalization Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a post sorting method, a device, equipment and a computer readable storage medium, wherein the method comprises the following steps: obtaining feature data for each post, wherein the feature data includes at least one of: detail page behavior data of a user in a detail page of each post, detail page content feature data of each post, and poster feature data of each post; calculating the point value of each post under each sorting index according to the feature data of each post; and sorting all the posts according to the point value of each post under each sorting index. The invention can eliminate posts with poor quality in the sequencing result and improve the experience degree of the user.
Description
Technical Field
The present invention relates to the field of internet technologies, and in particular, to a method, an apparatus, a device, and a computer-readable storage medium for sorting posts.
Background
With the continuous development of internet technology, data information in the internet is increasingly huge, and users often need to perform search operation to acquire desired information. In the prior art, the posts in the search result are usually sorted by using a single sort index, such as: the posts are sorted by time of posting or by click rate of the posts. However, the sorting mode in the prior art has the problems of single sorting index and poor sorting effect; moreover, the phenomenon that posts with poor quality appear in the sequencing result also exists, so that the user experience is influenced.
Disclosure of Invention
The embodiment of the invention mainly aims to provide a post sorting method, a device, equipment and a computer readable storage medium, which can eliminate posts with poor quality in a sorting result and improve the experience of a user.
In order to achieve the above object, an embodiment of the present invention provides a post ranking method, where the method includes:
obtaining feature data for each post, wherein the feature data includes at least one of: detail page behavior data of a user in a detail page of each post, detail page content feature data of each post, and poster feature data of each post;
calculating the point value of each post under each sorting index according to the feature data of each post;
and sorting all the posts according to the point value of each post under each sorting index.
Optionally, the detail page behavior data includes at least one of: click behavior data, browsing time data and input text data;
the detail page content characteristic data includes at least one of: post title characteristic information, post text characteristic information and post picture characteristic information;
the poster characteristic data comprises at least one of: the system comprises data of the identification grade of a poster, data of the online time length of the poster and data of the amount of the poster.
Optionally, the calculating, according to the feature data of each post, a score value of each post under each ranking index includes:
and selecting data required by an index training model from the feature data of each post, and training by using the index training model according to the selected data to obtain a score value of each post under a ranking index corresponding to the index training model.
Optionally, the sorting all the posts according to the score value of each post under each sorting index includes:
determining the lowest score value of each post under each sorting index according to the score value of each post under each sorting index;
and sorting all posts according to the determined lowest score value of each post.
Optionally, the sorting all the posts according to the score value of each post under each sorting index includes:
calculating the ranking of each post under one sorting index according to the score values of all posts under the one sorting index;
determining the lowest ranking of each post under all the ranking indexes according to the ranking of each post under each ranking index;
and sorting all the posts according to the determined lowest rank of each post.
Optionally, after sorting all posts, the method further comprises:
screening out n posts according to a sorting result and a preset screening rule, and displaying the n posts in a list page; wherein n is a positive integer.
Optionally, after sorting all posts, the method further comprises:
screening n posts according to a preset screening rule according to the sorting result, and calculating the sum of the score values of all sorting indexes of each screened post;
calculating a score value of a list page behavior index of each post after screening according to the list page behavior data of each post after screening in a list page by a user;
and sorting all the screened posts according to the product value of the score value of the list page behavior index of each screened post and the corresponding sum value, and displaying the sorting result in a list page.
In addition, to achieve the above object, an embodiment of the present invention further provides a post sorting apparatus, including:
an obtaining module, configured to obtain feature data of each post, where the feature data includes at least one of: detail page behavior data of a user in a detail page of each post, detail page content feature data of each post, and poster feature data of each post;
the calculation module is used for calculating the score value of each post under each sequencing index according to the feature data of each post;
and the ranking module is used for ranking all the posts according to the point value of each post under each ranking index.
In addition, to achieve the above object, an embodiment of the present invention further provides a post sorting apparatus, where the apparatus includes: a processor, a memory, and a communication bus;
the communication bus is used for realizing connection communication between the processor and the memory;
the processor is used for executing the post sorting program stored in the memory so as to realize the steps of the introduced post sorting method.
In addition, to achieve the above object, an embodiment of the present invention further provides a computer-readable storage medium, where a post ranking program is stored;
when executed by at least one processor, cause the at least one processor to perform the steps of the post ranking method described above.
The method, the device, the equipment and the computer-readable storage medium for sorting the posts, which are provided by the embodiment of the invention, generate the corresponding sorting index according to the post content and the poster information, so that the index types of the traditional user post sorting are enriched, the sorting effect of the posts is improved, poor posts do not exist in the sorting result, and the user experience is improved.
Drawings
FIG. 1 is a flow diagram of a post ranking method of a first embodiment of the present invention;
FIG. 2 is a flow chart of a post ranking method of a second embodiment of the present invention;
FIG. 3 is a flow chart of a post ranking method of a third embodiment of the present invention;
FIG. 4 is a flow chart of a post ranking method of a fourth embodiment of the present invention;
FIG. 5 is a flow chart of a post ranking method of a fifth embodiment of the present invention;
FIG. 6 is a diagram illustrating the structure of a post sorting device according to a sixth embodiment of the present invention;
fig. 7 is a schematic diagram showing the composition structure of a post ranking apparatus according to a seventh embodiment of the present invention.
Detailed Description
To further illustrate the technical means and effects of the embodiments of the present invention for achieving the intended purpose, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments.
A first embodiment of the present invention provides a method for sorting posts, as shown in fig. 1, the method specifically includes the following steps:
step S101: obtaining feature data for each post, wherein the feature data includes at least one of: detail page behavior data of the user in the detail page of each post, detail page content feature data of each post, and poster feature data of each post.
In the embodiment of the invention, the post is a record in a list page, and the list page comprises a plurality of posts; the detail page is the page entered after clicking on the post in the list page. For example, a search result page after searching on a search website is a list page, each search result presented in the list page is a post, and a page entered after clicking on a post is a detail page of the post.
Specifically, the detail page behavior data is data generated by the operation behavior of the user in the detail page of the post, and the detail page behavior data at least comprises one of the following data: click behavior data, browsing time data and input text data;
the detail page content feature data is data generated from the title and content of the post, the detail page content feature data including at least one of: post title characteristic information, post text characteristic information and post picture characteristic information;
the poster feature data is data generated from user information of posting posts, the poster feature data including at least one of: the system comprises data of the identification grade of a poster, data of the online time length of the poster and data of the amount of the poster.
Step S102: and calculating the point value of each post under each ranking index according to the feature data of each post.
In the embodiment, the point value of the post under the corresponding sorting index is calculated according to one or more types of data in the detail page behavior data, the detail page content characteristic data and the poster characteristic data of the post.
Step S103: and sorting all the posts according to the point value of each post under each sorting index.
And sorting all posts by utilizing a comprehensive quality sorting algorithm in the prior art according to the score value of each post under each sorting index.
In the embodiment of the invention, new characteristics which can be used for calculating the ranking index are mined from the content information and the poster information of the posts, so that the prediction effect of the original ranking index is improved; in addition, mining new features from the content information and poster information of posts can also be directly used as a ranking index for ranking posts.
A second embodiment of the present invention provides a method for sorting posts, as shown in fig. 2, the method specifically includes the following steps:
step S201: obtaining feature data for each post, wherein the feature data includes at least one of: detail page behavior data of the user in the detail page of each post, detail page content feature data of each post, and poster feature data of each post.
In the embodiment of the invention, the post is a record in a list page, and the list page comprises a plurality of posts; the detail page is the page entered after clicking on the post in the list page. For example, a search result page after searching on a search website is a list page, each search result presented in the list page is a post, and a page entered after clicking on a post is a detail page of the post.
Specifically, the detail page behavior data is data generated by the operation behavior of the user in the detail page of the post, and the detail page behavior data at least comprises one of the following data: click behavior data, browsing time data and input text data;
the detail page content feature data is data generated from the title and content of the post, the detail page content feature data including at least one of: post title characteristic information, post text characteristic information and post picture characteristic information;
the poster feature data is data generated from user information of posting posts, the poster feature data including at least one of: the system comprises data of the identification grade of a poster, data of the online time length of the poster and data of the amount of the poster.
Step S202: and calculating the point value of each post under each ranking index according to the feature data of each post.
In the embodiment, the point value of the post under the corresponding sorting index is calculated according to one or more types of data in the detail page behavior data, the detail page content characteristic data and the poster characteristic data of the post.
Specifically, step S202 includes:
and selecting data required by an index training model from the feature data of each post, and training by using the index training model according to the selected data to obtain a score value of each post under a ranking index corresponding to the index training model.
In the embodiment of the invention, a corresponding index training model is trained for each sequencing index in an off-line manner; during the online prediction, according to the requirement of each index training model, data required by the index training model is selected from the detail page behavior data, the detail page content characteristic data and the poster characteristic data of each post, and the score value of the post under the corresponding sorting index is trained according to the index training model.
Further, step S202 further includes:
normalizing the fraction value of each sequencing index into a certain numerical range according to the following formula:
wherein, Score is a Score value calculated according to the index training model;
mu is the average score value of all posts under any ranking index;
sigma is the standard deviation of all posts under any sort index;
Step S203: and determining the lowest score value of each post under each sorting index according to the score value of each post under each sorting index.
For example, if the score values of a post under each ranking index are: 0.384, -1.692, 0.854, -1.417, -0.640, the lowest score value of the post under each ranking indicator is-1.692.
Step S204: and sorting all posts according to the determined lowest score value of each post.
Specifically, after step S204, the method further includes:
screening out n posts according to a sorting result and a preset screening rule, and displaying the n posts in a list page; wherein n is a positive integer.
Further, if all posts are sorted in a descending order in step S203, a preset screening rule is to screen out the top n posts; if all posts are sorted in ascending order in step S203, the preset screening rule is to screen out the next n posts.
Compared with the prior art, the method and the device have the advantages that the point value of each post under a plurality of ranking indexes is calculated according to the new data mined from the content information and the poster information of the posts. All posts are sorted according to the minimum score value of the posts under a plurality of sorting indexes every day, and the posts with better quality are presented in the list page, so that the posts presented in the list page are all posts with higher quality, poor posts cannot exist, and the browsing experience of a user is improved.
A third embodiment of the present invention provides a method for sorting posts, as shown in fig. 3, the method specifically includes the following steps:
step S301: obtaining feature data for each post, wherein the feature data includes at least one of: detail page behavior data of the user in the detail page of each post, detail page content feature data of each post, and poster feature data of each post.
In the embodiment of the invention, the post is a record in a list page, and the list page comprises a plurality of posts; the detail page is the page entered after clicking on the post in the list page. For example, a search result page after searching on a search website is a list page, each search result presented in the list page is a post, and a page entered after clicking on a post is a detail page of the post.
Specifically, the detail page behavior data is data generated by the operation behavior of the user in the detail page of the post, and the detail page behavior data at least comprises one of the following data: click behavior data, browsing time data and input text data;
the detail page content feature data is data generated from the title and content of the post, the detail page content feature data including at least one of: post title characteristic information, post text characteristic information and post picture characteristic information;
the poster feature data is data generated from user information of posting posts, the poster feature data including at least one of: the system comprises data of the identification grade of a poster, data of the online time length of the poster and data of the amount of the poster.
Step S302: and calculating the point value of each post under each ranking index according to the feature data of each post.
In the embodiment, the point value of the post under the corresponding sorting index is calculated according to one or more types of data in the detail page behavior data, the detail page content characteristic data and the poster characteristic data of the post.
Specifically, step S302 includes:
and selecting data required by an index training model from the feature data of each post, and training by using the index training model according to the selected data to obtain a score value of each post under a ranking index corresponding to the index training model.
In the embodiment of the invention, a corresponding index training model is trained for each sequencing index in an off-line manner; during the online prediction, according to the requirement of each index training model, data required by the index training model is selected from the detail page behavior data, the detail page content characteristic data and the poster characteristic data of each post, and the score value of the post under the corresponding sorting index is trained according to the index training model.
Further, step S302 further includes:
normalizing the fraction value of each sequencing index into a certain numerical range according to the following formula:
wherein, Score is a Score value calculated according to the index training model;
mu is the average score value of all posts under any ranking index;
sigma is the standard deviation of all posts under any sort index;
Step S303: and determining the lowest score value of each post under each sorting index according to the score value of each post under each sorting index.
For example, if the score values of a post under each ranking index are: 0.384, -1.692, 0.854, -1.417, -0.640, the lowest score value of the post under each ranking indicator is-1.692.
Step S304: and sorting all posts according to the determined lowest score value of each post.
Step S305: and screening n posts according to a preset screening rule according to the sorting result, and calculating the sum of the score values of all the sorting indexes of each screened post.
Specifically, if all posts are sorted in descending order in step S304, the preset screening rule is to screen out the top n posts; if all posts are sorted in ascending order in step S304, the preset screening rule is to screen out the next n posts.
Step S306: and calculating the score value of the list page behavior index of each post after screening according to the list page behavior data of each post after screening in the list page by the user.
Specifically, the behavior data of the user for each post in the list page includes: click behavior data;
the list page behavior indicators include: click through rate.
Step S307: and sorting all the screened posts according to the product value of the score value of the list page behavior index of each screened post and the corresponding sum value, and displaying the sorting result in a list page.
In the embodiment of the invention, the posts are sorted twice, and preliminary sorting is carried out according to the lowest score of each post under each sorting index, so that part of posts with better quality are screened out from numerous posts; and secondly, performing secondary sorting on the posts by using the behavior data of each post in the list page of the user, so that the accuracy of sorting the posts can be improved, and the condition that the quality of the posts arranged in front is poor can be avoided.
A fourth embodiment of the present invention provides a post ranking method, as shown in fig. 4, the method specifically includes the following steps:
step S401: obtaining feature data for each post, wherein the feature data includes at least one of: detail page behavior data of the user in the detail page of each post, detail page content feature data of each post, and poster feature data of each post.
In the embodiment of the invention, the post is a record in a list page, and the list page comprises a plurality of posts; the detail page is the page entered after clicking on the post in the list page. For example, a search result page after searching on a search website is a list page, each search result presented in the list page is a post, and a page entered after clicking on a post is a detail page of the post.
Specifically, the detail page behavior data is data generated by the operation behavior of the user in the detail page of the post, and the detail page behavior data at least comprises one of the following data: click behavior data, browsing time data and input text data;
the detail page content feature data is data generated from the title and content of the post, the detail page content feature data including at least one of: post title characteristic information, post text characteristic information and post picture characteristic information;
the poster feature data is data generated from user information of posting posts, the poster feature data including at least one of: the system comprises data of the identification grade of a poster, data of the online time length of the poster and data of the amount of the poster.
Step S402: and calculating the point value of each post under each ranking index according to the feature data of each post.
In the embodiment, the point value of the post under the corresponding sorting index is calculated according to one or more types of data in the detail page behavior data, the detail page content characteristic data and the poster characteristic data of the post.
Specifically, step S402 includes:
and selecting data required by an index training model from the feature data of each post, and training by using the index training model according to the selected data to obtain a score value of each post under a ranking index corresponding to the index training model.
In the embodiment of the invention, a corresponding index training model is trained for each sequencing index in an off-line manner; during the online prediction, according to the requirement of each index training model, data required by the index training model is selected from the detail page behavior data, the detail page content characteristic data and the poster characteristic data of each post, and the score value of the post under the corresponding sorting index is trained according to the index training model.
Further, step S402 further includes:
normalizing the fraction value of each sequencing index into a certain numerical range according to the following formula:
wherein, Score is a Score value calculated according to the index training model;
mu is the average score value of all posts under any ranking index;
sigma is the standard deviation of all posts under any sort index;
Step S403: and calculating the ranking of each post under one sorting index according to the point values of all posts under the one sorting index.
In the embodiment of the present invention, the rank of each post under the index of the ranking is obtained according to the manner of step S403. For example, the ranking of a post under each ranking index is: 58. 12, 36, 41, 18.
Step S404: and determining the lowest ranking of each post under all the ranking indexes according to the ranking of each post under each ranking index.
The lowest ranking is the most lagging ranking, but is numerically the largest ranking. For example, if the ranks of a post under each ranking index are: 58. 12, 36, 41, 18, the lowest rank name of the post is 58.
Step S405: and sorting all the posts according to the determined lowest rank of each post.
Specifically, in the embodiment of the present invention, all posts are sorted in an ascending order according to the lowest rank of each post under each sorting index, so that the posts ranked in the front do not appear under the condition of extremely low rank under a certain sorting index, thereby ensuring that the comprehensive quality of the posts ranked in the front is high.
Specifically, after step S405, the method further includes:
screening out n posts according to a sorting result and a preset screening rule, and displaying the n posts in a list page; wherein n is a positive integer.
Further, if all posts are sorted in ascending order in step S405, the preset screening rule is to screen out the top n posts; if all posts are sorted in descending order in step S405, the preset screening rule is to screen out the next n posts.
Compared with the prior art, the method and the device have the advantages that the point value of each post under a plurality of ranking indexes is calculated according to the new data mined from the content information and the poster information of the posts. And sorting all the posts according to the point values of all the posts under the single sorting index to obtain the ranking of all the posts under the single sorting index. And then determining the lowest ranking of one post under a plurality of ranking indexes, namely the most lagging ranking, performing ascending ranking according to the lowest ranking of the posts, and displaying the posts ranked in the front in a list page. Therefore, the posts displayed in the list page are guaranteed to be high-quality posts, poor-quality posts cannot exist, and the browsing experience of the user is improved.
A fifth embodiment of the present invention provides a post ranking method, as shown in fig. 5, the method specifically includes the following steps:
step S501: obtaining feature data for each post, wherein the feature data includes at least one of: detail page behavior data of the user in the detail page of each post, detail page content feature data of each post, and poster feature data of each post.
In the embodiment of the invention, the post is a record in a list page, and the list page comprises a plurality of posts; the detail page is the page entered after clicking on the post in the list page. For example, a search result page after searching on a search website is a list page, each search result presented in the list page is a post, and a page entered after clicking on a post is a detail page of the post.
Specifically, the detail page behavior data is data generated by the operation behavior of the user in the detail page of the post, and the detail page behavior data at least comprises one of the following data: click behavior data, browsing time data and input text data;
the detail page content feature data is data generated from the title and content of the post, the detail page content feature data including at least one of: post title characteristic information, post text characteristic information and post picture characteristic information;
the poster feature data is data generated from user information of posting posts, the poster feature data including at least one of: the system comprises data of the identification grade of a poster, data of the online time length of the poster and data of the amount of the poster.
Step S502: and calculating the point value of each post under each ranking index according to the feature data of each post.
In the embodiment, the point value of the post under the corresponding sorting index is calculated according to one or more types of data in the detail page behavior data, the detail page content characteristic data and the poster characteristic data of the post.
Specifically, step S502 includes:
and selecting data required by an index training model from the feature data of each post, and training by using the index training model according to the selected data to obtain a score value of each post under a ranking index corresponding to the index training model.
In the embodiment of the invention, a corresponding index training model is trained for each sequencing index in an off-line manner; during the online prediction, according to the requirement of each index training model, data required by the index training model is selected from the detail page behavior data, the detail page content characteristic data and the poster characteristic data of each post, and the score value of the post under the corresponding sorting index is trained according to the index training model.
Further, step S502 further includes:
normalizing the fraction value of each sequencing index into a certain numerical range according to the following formula:
wherein, Score is a Score value calculated according to the index training model;
mu is the average score value of all posts under any ranking index;
sigma is the standard deviation of all posts under any sort index;
Step S503: and calculating the ranking of each post under one sorting index according to the point values of all posts under the one sorting index.
In the embodiment of the present invention, in the manner of step S503, the rank of a post under each rank as an index is obtained. For example, the ranking of a post under each ranking index is: 58. 12, 36, 41, 18.
Step S504: and determining the lowest ranking of each post under all the ranking indexes according to the ranking of each post under each ranking index.
The lowest ranking is the most lagging ranking, but is numerically the largest ranking. For example, if the ranks of a post under each ranking index are: 58. 12, 36, 41, 18, the lowest rank name of the post is 58.
Step S505: and sorting all the posts according to the determined lowest rank of each post.
Specifically, in the embodiment of the present invention, all posts are sorted in an ascending order according to the lowest rank of each post under each sorting index, so that the posts ranked in the front do not appear under the condition of extremely low rank under a certain sorting index, thereby ensuring that the comprehensive quality of the posts ranked in the front is high.
Step S506: and screening n posts according to a preset screening rule according to the sorting result, and calculating the sum of the score values of all the sorting indexes of each screened post.
Further, if all posts are sorted in ascending order in step S505, the preset screening rule is to screen out the top n posts; if all the posts are sorted in descending order in step S505, the preset screening rule is to screen out the next n posts.
Step S507: and calculating the score value of the list page behavior index of each post after screening according to the list page behavior data of each post after screening in the list page by the user.
Specifically, the behavior data of the user for each post in the list page includes: click behavior data;
the list page behavior indicators include: click through rate.
Step S508: and sorting all the screened posts according to the product value of the score value of the list page behavior index of each screened post and the corresponding sum value, and displaying the sorting result in a list page.
Preferably, all posts after the filtering are sorted in descending order in step S508.
In the embodiment of the invention, the posts are sorted twice, and the primary sorting is carried out according to the lowest ranking of each post under each sorting index, so that part of posts with better quality are screened out from numerous posts; and performing secondary sorting according to the list page behavior indexes of each screened post, thereby ensuring that the posts arranged in front do not have poor quality.
A sixth embodiment of the present invention provides a post sorting apparatus, as shown in fig. 6, the apparatus specifically includes the following components:
an obtaining module 601, configured to obtain feature data of each post, where the feature data includes at least one of: detail page behavior data of a user in a detail page of each post, detail page content feature data of each post, and poster feature data of each post;
a calculating module 602, configured to calculate, according to the feature data of each post, a score value of each post under each ranking index;
a sorting module 603, configured to sort all the posts according to the score value of each post under each sorting index.
Specifically, the detail page behavior data at least includes one of the following: click behavior data, browsing time data and input text data;
the detail page content characteristic data includes at least one of: post title characteristic information, post text characteristic information and post picture characteristic information;
the poster characteristic data comprises at least one of: the system comprises data of the identification grade of a poster, data of the online time length of the poster and data of the amount of the poster.
Further, the calculating module 602 is specifically configured to:
and selecting data required by an index training model from the feature data of each post, and training by using the index training model according to the selected data to obtain a score value of each post under a ranking index corresponding to the index training model.
Further, the sorting module 603 is specifically configured to:
determining the lowest score value of each post under each sorting index according to the score value of each post under each sorting index; sorting all posts according to the determined lowest point value of each post; or,
calculating the ranking of each post under one sorting index according to the score values of all posts under the one sorting index; determining the lowest ranking of each post under all the ranking indexes according to the ranking of each post under each ranking index; and sorting all the posts according to the determined lowest rank of each post.
Still further, the apparatus further comprises:
the display module is used for screening out n posts according to the sorting result and a preset screening rule and displaying the n posts in a list page; wherein n is a positive integer; or,
the system is used for screening n posts according to the sorting result and a preset screening rule and calculating the sum of the scores of all sorting indexes of each screened post; calculating a score value of a list page behavior index of each post after screening according to the list page behavior data of each post after screening in a list page by a user; sorting all the screened posts according to the product value of the score value of the list page behavior index of each screened post and the corresponding sum value, and displaying the sorting result in the list page
A seventh embodiment of the present invention proposes a post ranking apparatus, as shown in fig. 7, the apparatus comprising: a processor 701, a memory 702, and a communication bus;
the communication bus is used for realizing connection communication between the processor 701 and the memory 702;
processor 701 is configured to execute a post ranking program stored in memory 702 to implement the following steps:
obtaining feature data for each post, wherein the feature data includes at least one of: detail page behavior data of a user in a detail page of each post, detail page content feature data of each post, and poster feature data of each post;
calculating the point value of each post under each sorting index according to the feature data of each post;
and sorting all the posts according to the point value of each post under each sorting index.
An eighth embodiment of the present invention provides a computer-readable storage medium storing a post ranking program;
when executed by at least one processor, the post ranking program causes the at least one processor to perform the steps of:
obtaining feature data for each post, wherein the feature data includes at least one of: detail page behavior data of a user in a detail page of each post, detail page content feature data of each post, and poster feature data of each post;
calculating the point value of each post under each sorting index according to the feature data of each post;
and sorting all the posts according to the point value of each post under each sorting index.
The method, the device, the equipment and the computer-readable storage medium for sorting the posts generate the corresponding sorting indexes according to the post content and the poster information, so that the index types of the traditional user post sorting are enriched, the sorting effect of the posts is improved, poor posts do not exist in the sorting result, and the user experience is improved.
While the embodiments of the present invention have been described in terms of specific embodiments, it is to be understood that both the foregoing general description and the following detailed description are intended to provide further explanation and understanding of the invention as claimed.
Claims (9)
1. A method of ranking posts, the method comprising:
obtaining feature data for each post, wherein the feature data includes at least one of: detail page behavior data of a user in a detail page of each post, detail page content feature data of each post, and poster feature data of each post;
calculating the point value of each post under each sorting index according to the feature data of each post;
sorting all posts according to the point value of each post under each sorting index;
the ranking all posts according to the point value of each post under each ranking index comprises the following steps:
determining the lowest score value of each post under each sorting index according to the score value of each post under each sorting index;
and sorting all posts according to the determined lowest score value of each post.
2. The method of claim 1, wherein the detail page behavior data comprises at least one of: click behavior data, browsing time data and input text data;
the detail page content characteristic data includes at least one of: post title characteristic information, post text characteristic information and post picture characteristic information;
the poster characteristic data comprises at least one of: the system comprises data of the identification grade of a poster, data of the online time length of the poster and data of the amount of the poster.
3. The method of claim 1, wherein the calculating a score value for each post at a respective ranking indicator based on the feature data of each post comprises:
and selecting data required by an index training model from the feature data of each post, and training by using the index training model according to the selected data to obtain a score value of each post under a ranking index corresponding to the index training model.
4. The method of claim 1, wherein the ranking all posts according to the value of each post's score under a respective ranking indicator comprises:
calculating the ranking of each post under one sorting index according to the score values of all posts under the one sorting index;
determining the lowest ranking of each post under all the ranking indexes according to the ranking of each post under each ranking index;
and sorting all the posts according to the determined lowest rank of each post.
5. The method of sorting posts of any of claims 1-4, wherein after sorting all posts, the method further comprises:
screening out n posts according to a sorting result and a preset screening rule, and displaying the n posts in a list page; wherein n is a positive integer.
6. The method of sorting posts of any of claims 1-4, wherein after sorting all posts, the method further comprises:
screening n posts according to a preset screening rule according to the sorting result, and calculating the sum of the score values of all sorting indexes of each screened post;
calculating a score value of a list page behavior index of each post after screening according to the list page behavior data of each post after screening in a list page by a user;
and sorting all the screened posts according to the product value of the score value of the list page behavior index of each screened post and the corresponding sum value, and displaying the sorting result in a list page.
7. A post ranking apparatus, the apparatus comprising:
an obtaining module, configured to obtain feature data of each post, where the feature data includes at least one of: detail page behavior data of a user in a detail page of each post, detail page content feature data of each post, and poster feature data of each post;
the calculation module is used for calculating the score value of each post under each sequencing index according to the feature data of each post;
the ranking module is used for ranking all posts according to the point value of each post under each ranking index;
the sorting module is specifically configured to: determining the lowest score value of each post under each sorting index according to the score value of each post under each sorting index;
and sorting all posts according to the determined lowest score value of each post.
8. A post ranking apparatus, the apparatus comprising: a processor, a memory, and a communication bus;
the communication bus is used for realizing connection communication between the processor and the memory;
the processor is configured to execute a post ranking program stored in the memory to implement the steps of the post ranking method of any of claims 1 to 6.
9. A computer-readable storage medium, wherein the computer-readable storage medium stores a post ranking program;
the post ranking program, when executed by at least one processor, causes the at least one processor to perform the steps of the post ranking method of any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810597468.1A CN108717469B (en) | 2018-06-11 | 2018-06-11 | Post sorting method, device and equipment and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810597468.1A CN108717469B (en) | 2018-06-11 | 2018-06-11 | Post sorting method, device and equipment and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108717469A CN108717469A (en) | 2018-10-30 |
| CN108717469B true CN108717469B (en) | 2021-11-23 |
Family
ID=63911885
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810597468.1A Active CN108717469B (en) | 2018-06-11 | 2018-06-11 | Post sorting method, device and equipment and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108717469B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109597941B (en) * | 2018-12-12 | 2021-04-27 | 拉扎斯网络科技（上海）有限公司 | Sorting method and device, electronic equipment and storage medium |
| CN112115334B (en) * | 2020-09-28 | 2023-07-21 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for distinguishing network community hot content |
| CN112765346B (en) * | 2020-11-18 | 2022-10-28 | 北京五八信息技术有限公司 | Information processing method and device |
| CN113496005B (en) * | 2021-05-26 | 2022-04-08 | 北京房多多信息技术有限公司 | Information management method and device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8751917B2 (en) * | 2011-11-30 | 2014-06-10 | Facebook, Inc. | Social context for a page containing content from a global community |
| CN105917364A (en) * | 2013-12-31 | 2016-08-31 | 微软技术许可有限责任公司 | Ranking of discussion threads in a question-and-answer forum |
| CN106886561A (en) * | 2016-12-29 | 2017-06-23 | 中国科学院自动化研究所 | Web Community's model influence sort method based on association in time interaction fusion |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070160963A1 (en) * | 2006-01-10 | 2007-07-12 | International Business Machines Corporation | Candidate evaluation tool |
-
2018
- 2018-06-11 CN CN201810597468.1A patent/CN108717469B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8751917B2 (en) * | 2011-11-30 | 2014-06-10 | Facebook, Inc. | Social context for a page containing content from a global community |
| CN105917364A (en) * | 2013-12-31 | 2016-08-31 | 微软技术许可有限责任公司 | Ranking of discussion threads in a question-and-answer forum |
| CN106886561A (en) * | 2016-12-29 | 2017-06-23 | 中国科学院自动化研究所 | Web Community's model influence sort method based on association in time interaction fusion |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108717469A (en) | 2018-10-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108717469B (en) | Post sorting method, device and equipment and computer readable storage medium | |
| CN109145216B (en) | Network public opinion monitoring method, device and storage medium | |
| US8667037B1 (en) | Identification and ranking of news stories of interest | |
| CN108460082B (en) | Recommendation method and device and electronic equipment | |
| JP2017515249A5 (en) | ||
| CN102708168A (en) | System and method for sorting search results of teaching resources | |
| CN105930469A (en) | Hadoop-based individualized tourism recommendation system and method | |
| TWI525456B (en) | Choose font, font determination, recommendation, generation method and device | |
| CN111858905B (en) | Model training method, information identification device, electronic equipment and storage medium | |
| US20150254574A1 (en) | Related data generating apparatus, related data generating method, and program | |
| US20180189291A1 (en) | Method and apparatus for sorting related searches | |
| CN112825089B (en) | Article recommendation method, device, equipment and storage medium | |
| US10073891B2 (en) | Forensic system, forensic method, and forensic program | |
| CN102957949A (en) | Device and method for recommending video to user | |
| CN108804676B (en) | Post sorting method, device and equipment and computer readable storage medium | |
| CN112667571A (en) | Biomedical literature search and sorting method and device | |
| CN103106234A (en) | Searching method and device of webpage content | |
| CN117726311B (en) | Intelligent matching method, device, equipment and storage medium for employment posts of supervision objects | |
| AU2013201006B2 (en) | Information classification program, information classification method, and information processing apparatus | |
| CN112612961B (en) | Information searching method, device, storage medium and computer equipment | |
| US20140156693A1 (en) | Filmstrip-based query suggestions | |
| US20100057724A1 (en) | Server device for creating list of general words to be excluded from search result | |
| US20150193444A1 (en) | System and method to determine social relevance of Internet content | |
| CN108804674B (en) | Post sorting method, device and equipment and computer readable storage medium | |
| CN110941709A (en) | Information screening method and device, electronic equipment and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |