CN112948706B - Article generation method and device based on comment recommendation and storage medium - Google Patents
Article generation method and device based on comment recommendation and storage medium Download PDFInfo
- Publication number
- CN112948706B CN112948706B CN202110083047.9A CN202110083047A CN112948706B CN 112948706 B CN112948706 B CN 112948706B CN 202110083047 A CN202110083047 A CN 202110083047A CN 112948706 B CN112948706 B CN 112948706B
- Authority
- CN
- China
- Prior art keywords
- comment
- comments
- recommendation
- atp
- review
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/34—Browsing; Visualisation therefor
- G06F16/345—Summarisation for human users
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a comment recommendation-based article generation method and device and a storage medium. The article generation method based on comment recommendation comprises the following steps: performing recommendation operation on the comments; inputting a recommended title of the comment; submitting a recommendation record of the comment; auditing the recommendation records based on the comments; if the audit on the recommended record is not passed, setting the recommended record state as not passed; if the review of the recommended record is passed, entering the next step; entering the comments as a new article; adding a user selected topic to the new article; setting the state of the new article as pass; a summary of the new article is displayed in the article list. When the old article carries the high-quality comments, the comments are prevented from being buried along with the old article, so that the comments are discussed further, and the value of the comments is fully mined.
Description
Technical Field
The invention relates to the field of Internet, in particular to a comment recommendation-based article generation method and device and a storage medium.
Background
The internet produces a large number of articles and reviews each day. Without the lack of quality comments. These comments are of great value in discussion. However, new articles are constantly being generated. The old article quickly became overwhelmed. Meanwhile, comments are buried with old articles. The value of the comments is not fully mined. Accordingly, those skilled in the art have endeavored to develop a review recommendation-based article generation method, apparatus, and storage medium.
Disclosure of Invention
In view of the above-mentioned drawbacks of the prior art, the technical problem to be solved by the present invention is how to achieve sufficient mining of the value of the comments.
In order to achieve the purpose, the invention provides a comment recommendation-based article generation method, a comment recommendation-based article generation device and a storage medium.
In an embodiment of the invention, an article generation method based on comment recommendation comprises the following steps:
performing recommendation operation on the comments;
inputting a recommended title of the comment;
submitting a recommendation record of the comment;
auditing the recommendation record based on the review;
if the audit on the recommendation record is not passed, setting the state of the recommendation record as not passed;
if the review of the recommended record is passed, entering the next step;
entering the comments as a new article;
adding a user selected topic to the new article;
setting the state of the new article as pass;
displaying the abstract of the new article in an article list.
In another embodiment of the invention, an article generation device based on comment recommendation comprises a foreground module and a background module;
the foreground module carries out recommendation operation on the comments; inputting a recommended title of the comment;
the background module submits a recommendation record of the comment; auditing the recommendation record based on the review; if the audit on the recommendation record is not passed, setting the state of the recommendation record as not passed; if the review of the recommended record is passed, entering the next step; entering the comments as a new article; adding a user selected topic to the new article; setting the state of the new article as pass;
the foreground module displays the summaries of the new articles in an article list.
In another embodiment of the invention, a computer-readable storage medium comprises a computer program that runs on a computer, which executes the method.
The comment recommendation-based article generation method, the comment recommendation-based article generation device and the storage medium have the following beneficial effects: when the old article carries a high-quality comment, performing recommendation operation on the comment; the comment is recorded as a new article, so that the comment is prevented from being buried along with the old article; adding a user selected topic to the new article based on the comment so that the comment can be discussed further; a summary of a new article based on the review is displayed in the article list, enabling full mining of the value of the review.
The conception, specific structure and technical effects of the present invention will be further described in conjunction with the accompanying drawings to fully understand the purpose, characteristics and effects of the present invention.
Drawings
FIG. 1 is a flow diagram illustrating an embodiment of a method for article generation based on comment recommendation in accordance with the present invention;
FIGS. 2 and 3 are schematic diagrams of user interfaces of an embodiment of a comment recommendation-based article generation method according to the present invention;
FIG. 4 is a flowchart illustrating another embodiment of a method for article generation based on comment recommendation according to the present invention;
FIG. 5 is a schematic diagram of a user interface of another embodiment of a method for article generation based on comment recommendation of the present invention;
FIG. 6 is a flowchart illustrating another embodiment of a method for article generation based on comment recommendation according to the present invention;
FIG. 7 is a flowchart illustrating a method for generating a comment-based recommended article according to another embodiment of the present invention;
FIG. 8 is a block diagram of an embodiment of an article generation apparatus for comment-based recommendation in accordance with the present invention;
FIG. 9 is a block diagram of an article generation apparatus based on comment recommendation according to another embodiment of the present invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the drawings only show the components related to the present invention rather than the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.
Some exemplary embodiments of the invention have been described for illustrative purposes, and it is to be understood that the invention may be practiced otherwise than as specifically described.
FIG. 1 is a flow diagram illustrating an embodiment of a method for article generation based on comment recommendation in accordance with the present invention; fig. 2 and 3 are schematic user interfaces of an embodiment of an article generation method based on comment recommendation according to the present invention.
As shown in fig. 1 to 3, the article generation method 100 based on comment recommendation includes steps S110 to S190.
In step S110, a recommendation operation is performed for the comment.
First, the website publishes old articles. Reviewers have browsed older articles and have published quality reviews. The recommender browses the comment, and performs a recommendation operation.
Step S120, a recommended title of the comment is input.
The website asks the recommender for the recommended title of the comment. The recommender inputs the recommended title of the above comments to the website.
And step S130, submitting the recommendation record of the comment.
And the website prompts the recommender to submit the recommendation record of the comments. And the recommender submits the recommendation record of the comments to the website.
And step S140, auditing the recommendation records based on the comments.
And the website displays the recommendation records of the comments to an administrator. And the administrator browses the recommendation records of the comments on the website and feeds back the audit result.
Step S150A, if the audit on the recommendation record is not passed, setting the state of the recommendation record as not passed.
And the administrator considers that the recommended record has no value, the feedback of the failure of the audit is sent to the website, and the website sets the recommended record to be in the failure state according to the audit of the administrator.
And step S150B, if the audit on the recommended record is passed, the next step is carried out.
And the administrator considers that the recommended records have value, the approval passing is fed back to the website, and the website executes the subsequent steps according to the approval of the administrator.
Step S160, the comment is entered as a new article.
And the content of the comments is recorded as a new article when the audit is passed. The new articles are saved in an article list.
Step S170, adding the user' S selected topics to the new article.
The new article may add a user selected topic. Such new articles are at the front of the article list.
Step S180, the state of the new article is set to pass.
The contents of the above comments have been reviewed. New articles based on the above comments also pass.
In step S190A, a summary of the new article is displayed in the article list.
The website may display a list of articles and summaries of their articles. The reader can easily browse from them new articles based on the above comments.
In the method provided by the embodiment, when the old article carries a high-quality comment, recommendation operation is carried out on the comment; the comment is recorded as a new article, so that the comment is prevented from being buried along with the old article; adding a user selected topic to the new article based on the comment so that the comment can be discussed further; a summary of a new article based on the review is displayed in the article list, enabling full mining of the value of the review.
In another embodiment, the website publishes old articles. The first reviewer browses the old article and posts the first review. And a first recommender carries out recommendation operation on the first comment through the website, inputs a recommendation title of the first comment, and submits a recommendation record of the first comment. The administrator of the website audits the recommendation record based on the first comment, and if the audit of the recommendation record based on the first comment is not passed, the state of the recommendation record is set to be not passed; if the review for the recommendation record based on the first review passes, then the next step is entered. The website inputs the first comment into a first new article, adds a user selected topic to the first new article, sets the state of the first new article to be a passing state, and displays an abstract of the first new article in an article list. The second reviewer browses the old article and posts a second review. And the second recommender carries out recommendation operation on the second comment through the website, inputs the recommendation title of the second comment and submits the recommendation record of the second comment. The administrator of the website audits the recommended record based on the second comment, and if the audit of the recommended record based on the second comment is not passed, the recommended record is set to be in a non-pass state; if the review for the recommendation record based on the second comment passes, the next step is entered. The website enters the second comment into a second new article, adds a user selected topic to the second new article, sets the state of the second new article to be a passing state, and displays the abstract of the second new article in the article list. Thereafter, the above steps are repeated in a loop.
In this way, the first comment with high quality is recommended and then entered as a first new article. More users see the first new article and continue to discuss producing a second comment of good quality. And the second comment is recommended and then is recorded as a second new article. Such a repetitive cycle allows the high-quality reviews to be fully discussed and their value to be fully explored.
In another embodiment, the website publishes the old article of the author. The reviewer browses the old article and published the first review. The author browses the first comment and replies to the second comment. And the recommender carries out recommendation operation on the first comment or the second comment through the website, inputs the recommendation title of the first comment or the second comment, and submits the recommendation record of the first comment or the second comment. The administrator of the website audits the recommendation record based on the first comment or the second comment, and if the audit of the recommendation record based on the first comment or the second comment is not passed, the state of the recommendation record is set as not passed; and if the review for the recommendation record based on the first comment or the second comment passes, entering the next step. The website inputs the first comment or the second comment into a new article, adds a user selected topic to the new article, sets the state of the new article to be passed, and displays the abstract of the new article in an article list.
In this way, the first comment posted by the reviewer may be entered as a new article via the recommendation. The second comment replied by the author may also be entered as a new article via a recommendation. Both the first comment and the second comment can be discussed sufficiently and their value can be exploited sufficiently.
FIG. 4 is a flowchart illustrating a method for generating a comment-based recommended article according to another embodiment of the present invention; FIG. 5 is a schematic diagram of a user interface of another embodiment of an article generation method based on comment recommendation according to the present invention.
As shown in fig. 4 and 5, the article generation method 100 based on comment recommendation further includes step S182 to step S190B.
In step S182, after the new article is set to pass, the identification number of the new article is generated.
The content of the quality comments has been approved. New articles based on the above comments are passed. The website then generates an identification number for the new article. The identification number of the new article is unique.
And step S184, recording the identification number of the new article into the recommendation record.
The identification number of the new article is entered into the recommendation record of the above comments. The new article, the comments and the recommendation record are associated with each other.
In step S186, a comment is displayed in the comment recommendation area.
The website is provided with a comment recommendation area under the old article. The comment recommendation field displays the above comments to the reader. The reader can browse the comments in the comment recommendation area.
In step S188, when the comment is operated, the identification number of the new article is acquired from the recommendation record.
The comments in the comment recommendation area are provided with a label of 'who recommends'. The reader can click on the "who recommended" tab on the review. At this time, the website acquires the identification number of the new article from the recommendation record of the comment. The website can acquire the full text of the new article according to the identification number of the new article.
And step S190B, displaying the full text of the new article.
The website may display a full text of a new article based on the comments. The reader can easily browse the full text of the new article from it. In addition, the website may display a list of articles and summaries of their articles. The reader can easily browse the abstract of the new article from it.
The method of the embodiment displays the full text and abstract of the new article in different ways. The reader may conveniently choose to view the full text or abstract of the new article.
FIG. 6 is a flowchart illustrating an article generating method based on comment recommendation according to another embodiment of the present invention.
As shown in fig. 6, the article generation method 100 based on comment recommendation further includes steps S142A to S148A.
And step S142A, counting the times of approval of the comments on the current day and the past day after auditing the recommendation records based on the comments.
Step S144A, calculate the first heat of the comment according to the formula B (W) = (Atp + Tp × R) ÷ ((Btp + Atp) + Tp) of the first algorithm, where B (W) is the first heat of the comment, atp is the number of times the comment is approved on the current day, btp is the number of times the comment is approved on the past day, btp, atp sum is the total number of times the comment is approved, tp is the ratio of the sum of Btp, atp of all comments to the number thereof, and R is the average of the ratio of Atp of all comments to the sum of Btp, atp thereof.
Step S146A, determining whether the comment is a hotspot comment according to the first popularity of the comment.
In step S148A, if the comment is a hotspot comment, the review of the recommendation record is passed.
The embodiment adopts a first algorithm to calculate the first popularity of the comment after the comment is recommended. The first algorithm eliminates the effect of too few numbers of all reviews on the popularity of the review. And recommending that the record is approved when the first popularity of the comment reaches the standard of the hot comment. Thus, review of the review-based recommendation record is automatically completed without manual action.
FIG. 7 is a flowchart illustrating a method for generating a comment-based recommended article according to another embodiment of the present invention.
As shown in fig. 7, the article generation method 100 based on comment recommendation further includes steps S142B to S148B.
And step S142B, counting the times of approval of the comments on the current day and the past day after auditing the recommendation records based on the comments.
Step S144B, according to the formula of the second algorithm
Calculating a second degree of hotness of the comment, wherein
For the second popularity of the review, atp is the number of times the review was approved on the current day, and Btp is the number of times the review was approved on the past day.
And step S146B, judging whether the comment is a hotspot comment according to the second popularity of the comment.
In step S148B, if the comment is a hot comment, the review for the recommendation record is passed.
The embodiment calculates the second popularity of the comment by adopting a second algorithm after the comment is recommended. The second algorithm corrects the variation of the popularity of the review over time after it is posted. And the second popularity of the comment meets the standard of the hotspot comment, and the record is recommended to be approved. Thus, review of the review-based recommendation record is automatically completed without manual action.
In another embodiment, the article generation method based on comment recommendation 100 further includes:
after reviewing the review-based recommendation record,
counting the number of times that the comment is praised on the current day and the past day;
calculating a first popularity of the comment according to a formula B (W) = (Atp + Tp × R) ÷ ((Btp + Atp) + Tp) of a first algorithm, wherein B (W) is the first popularity of the comment, atp is the number of times the comment is liked on the current day, btp is the number of times the comment is liked on the previous day, btp, atp sum is the total number of times the comment is liked, tp is the ratio of the total number of Btp, atp and the number thereof of all comments, and R is an average value of the ratio of Atp and the total number of Btp, atp thereof of all comments;
formula according to the second algorithm
Calculating a second degree of hotness of the comment, wherein
For the second popularity of the comment, atp is the number of times the comment is liked on the current day, and Btp is the number of times the comment is liked on the past day;
according to the formula of weighted averaging
Calculating a weighted popularity of the comment, wherein H (W) is the weighted popularity of the comment, α is a first weight, and β is a second weight;
judging whether the comment is a hot comment or not according to the weighted heat of the comment;
if the review is a hotspot review, the review for the recommendation record passes.
The embodiment calculates the first popularity of the comment by adopting a first algorithm after the comment is recommended. The first algorithm eliminates the effect of too few numbers of all reviews on the popularity of the review. Here, a second algorithm is also used to calculate a second popularity of the review after the review recommendation. The second algorithm corrects the variation of the popularity of the review over time after it is posted. Here, the weighted heat of the comment is also calculated by a weighted average method. The weighted average method has the advantages of both the first algorithm and the second algorithm. The weighted heat degree of the comment more accurately reflects the fire heat degree of the comment.
In another embodiment, the article recommendation based on comment generating method 100 further comprises:
after reviewing the recommendation records based on the comments, counting the number of times that the comments are stepped on the current day and the past day;
calculating a first negative heat degree of the comment according to a formula B (W) = (Atp + Tp × R) ÷ ((Btp + Atp) + Tp) of a first algorithm, wherein B (W) is the first negative heat degree of the comment, atp is the number of times the comment is stepped on the day, btp is the number of times the comment is stepped on the past day, the total of Btp and Atp is the total number of times the comment is stepped on the past day, tp is the ratio of the total of Btp and Atp to the number of the comments, and R is an average value of the ratio of Atp of all comments to the total of Btp and Atp of the comments;
judging whether the comment is a negative comment according to the first negative heat of the comment;
if the review is a negative review, the review for the recommendation record fails.
The embodiment adopts a first algorithm to calculate the first negative heat of the comment after the comment is recommended. The first algorithm eliminates the effect of too few numbers of all reviews on the negative popularity of the review. The first negative heat of the review meets the criteria for a negative review recommends that the record review failed. Thus, review of the review-based recommendation record is automatically completed without manual action.
In another embodiment, the article generation method based on comment recommendation 100 further includes:
after the recommendation records based on the comments are audited, counting the number of times that the comments are stepped on the current day and the past day;
formula according to the second algorithm
Calculating a second negative heat of the review, wherein
For the second negative heat of the comment, atp is the number of times the comment is stepped on the current day, and Btp is the number of times the comment is stepped on the past day;
judging whether the comment is a negative comment according to the second negative heat of the comment;
if the review is a negative review, the review for the recommendation record fails.
The embodiment adopts a second algorithm to calculate the second negative heat of the comment after the comment is recommended. The second algorithm corrects the negative heat over time after the review is posted. The second negative heat of the review meets the criteria for a negative review recommends that the record review failed. Thus, review of the review-based recommendation record is automatically completed without manual operation.
In another embodiment, the article recommendation based on comment generating method 100 further comprises:
after reviewing the review-based recommendation record,
counting the number of times that the comment is praised on the current day and the past day;
formula according to the third algorithm Score = (P-1) ÷ (T + 2) G Calculating a third popularity of the comment, wherein Score is the third popularity of the comment, P is a total number of praise for the comment, T is a time until the comment appears, G is a gravity factor, and a default value of G is 1.8;
judging whether the comment is a hot comment or not according to the third popularity of the comment;
if the review is a hotspot review, the review for the recommendation record passes.
The present embodiment calculates the third popularity of the comment by using the third algorithm after the comment recommendation. The third algorithm derives the degree of fire heat for the review, primarily based on the total number of endorsements for the review. And recommending that the record passes the review when the third popularity of the comment reaches the standard of the hot comment. Thus, review of the review-based recommendation record is automatically completed without manual operation.
In another embodiment, the article generation method based on comment recommendation 100 further includes:
after reviewing the review-based recommendation record,
counting the times of praise and trample of the comment on the current day and the past day;
the formula Score = ㏒ according to the fourth algorithm 10 The fourth heat of the comment is calculated as z + yt/45000, where Score is the fourth heat of the comment, z is the difference between the number of times the comment is voted and the number of times the comment is stepped on, t is the time when the comment is posted, y is the voting direction of the comment, y =1 when z > 0, y =0 when z =0, and y = -1 when z < 0;
judging whether the comment is a hot comment according to the fourth popularity of the comment;
if the review is a hotspot review, the review for the recommendation record passes.
The fourth algorithm is adopted by the embodiment to calculate the fourth popularity of the comment after the comment is recommended. The fourth algorithm is mainly determined according to the difference between the number of times the comment is approved and the number of times the comment is stepped on. And recommending that the record passes the review when the fourth popularity of the comment reaches the standard of the hot comment. Thus, review of the review-based recommendation record is automatically completed without manual action.
FIG. 8 is a block diagram of an article generating apparatus for comment-based recommendation according to an embodiment of the present invention.
As shown in fig. 8, the article generation apparatus 200 based on comment recommendation includes a foreground module 210, a background module 220;
the foreground module 210 performs recommendation operation on the comments; inputting a recommended title of the comment;
the background module 220 submits a recommendation record of the comment; auditing the recommendation records based on the comments; if the audit on the recommended record is not passed, setting the recommended record state as not passed; if the recommended record is approved, entering the next step; entering the comments as a new article; adding a user selected topic to the new article; setting the state of the new article as pass;
the foreground module 210 displays the summaries of the new articles in the article list.
The device provided by the embodiment is applied to the website. When the old article carries a high-quality comment, performing recommendation operation on the comment; the comment is recorded as a new article, so that the comment is prevented from being buried along with the old article; adding a user selected topic to the new article based on the comment so that the comment can be discussed further; a summary of a new article based on the review is displayed in the article list, enabling full mining of the value of the review.
As shown in fig. 8, after the background module 220 sets the state of the new article as pass, it generates an identification number of the new article; recording the identification number of the new article into a recommendation record;
the foreground module 210 displays the comment in the comment recommendation area;
when the backstage module 220 operates on the comments, the identification number of the new article is obtained from the recommendation record;
the foreground module 210 displays the full text of the new article.
The apparatus of the present embodiment can display the full text of a new article based on the above comments. The reader can easily browse the full text of the new article from there. In addition, the apparatus herein may display a list of articles and summaries of the articles thereof. The reader can easily browse the abstract of the new article from it.
The device of the embodiment displays the full text and the abstract of the new article in different modes. The reader may conveniently choose to view the full text or abstract of the new article.
As shown in fig. 8, the back-office module 220, after reviewing the review-based recommendation record,
counting the number of times that the comment is approved on the current day and the past day;
calculating a first popularity of the comment according to a formula B (W) = (Atp + Tp × R) ÷ ((Btp + Atp) + Tp) of a first algorithm, wherein B (W) is the first popularity of the comment, atp is the number of times the comment is liked on the current day, btp is the number of times the comment is liked on the previous day, btp, atp sum is the total number of times the comment is liked, tp is the ratio of the total number of Btp, atp and the number thereof of all comments, and R is an average value of the ratio of Atp and the total number of Btp, atp thereof of all comments;
judging whether the comment is a hot comment or not according to the first popularity of the comment;
if the review is a hotspot review, the review for the recommendation record passes.
As shown in fig. 8, the back-office module 220, after reviewing the review-based recommendation record,
counting the number of times that the comment is praised on the current day and the past day;
formula according to the second algorithm
Calculating a second degree of hotness of the comment, wherein
For the second popularity of the comment, atp is the number of times the comment is liked on the current day, and Btp is the number of times the comment is liked on the past day;
judging whether the comment is a hot comment or not according to the second popularity of the comment;
if the review is a hotspot review, the review for the recommendation record passes.
FIG. 9 is a block diagram of an article generating apparatus based on comment recommendation according to another embodiment of the present invention.
As shown in fig. 9, the article generation apparatus 200 based on comment recommendation further includes a database module 230;
the background module 220 enters information such as a recommendation title of the comment into the recommendation record.
As shown in Table 1, the recommendation record for database module 230 includes a number of fields: id. user _ id, comment _ id, created _ at, and recimmend _ title. The first field id is number, type int, length 11. The second field, user _ id, is recommender number, type int, length 11. The third field comment _ id is the reviewer number, type int, length 11. The fourth field created _ at is the comment creation time, of type timestamp, and of length 0. The fifth field, recimmend _ title, is the recommended title, type varchar, length 100.
TABLE 1
| Name (name) | Type (B) | Length of | |
| id | int | 11 | |
| user_id | int | 11 | |
| comment_id | int | 11 | |
| created_at | timestamp | 0 | |
| | varchar | 100 |
As shown in Table 2, database module 230 maintains a plurality of recommendation records. Record 1 has id 102, user \/id 319, comment \/id 25615217, created _at2020-08-17, 41, and recimmend _ title is "band summer made … …". The 2 nd recommended record has id 103, user \/u id 8504, comment \/u id 25612292, created \/u at 2020-08-17, 32, and recommend _ title "why no … …. Entry 3 recommended the id of 104, user _idof 120770, comment _idof 25602703, created _atof 2020-08-17 34, and recimmend _ title of "church-river under-the-sky … ….
TABLE 2
The present invention also provides a computer-readable storage medium including a computer program, the computer program being run on a computer, the computer executing the article generating method recommended based on comments. The embodiment of the article generation method based on comment recommendation is described above, and is not repeated.
In summary, the invention discloses a comment recommendation-based article generation method, a comment recommendation-based article generation device and a storage medium. The article generation method based on comment recommendation comprises the following steps: performing recommendation operation on the comments; inputting a recommended title of the comment; submitting a recommendation record of the comment; auditing the recommendation records based on the comments; if the audit on the recommended record is not passed, setting the recommended record state as not passed; if the review of the recommended record is passed, entering the next step; entering the comments as a new article; adding a user selected topic to the new article; setting the state of the new article as pass; a summary of the new article is displayed in the article list. When the old article carries the high-quality comments, the comments are prevented from being buried along with the old article, so that the comments are discussed further, and the value of the comments is fully mined.
The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.
Claims (4)
1. An article generation method based on comment recommendation is characterized by comprising the following steps: performing recommendation operation on the comments; inputting a recommended title of the comment; submitting a recommendation record of the comment; auditing the recommendation record based on the review; if the audit on the recommendation record is not passed, setting the state of the recommendation record as not passed; if the review of the recommended record is passed, entering the next step; entering the comments as a new article; adding a user selected topic to the new article; setting the state of the new article as pass; displaying the abstract of the new article in an article list;
after the new article is set to pass, generating an identification number of the new article; recording the identification number of the new article into the recommendation record; displaying the comment in a comment recommendation area; when the comment is operated, acquiring the identification number of the new article from the recommendation record; displaying the full text of the new article;
after reviewing recommendation records based on comments, counting the number of times that the comments are approved on the current day and the past day, and calculating a first heat of the comments according to a formula B (W) = (Atp + Tp × R) ÷ ((Btp + Atp) + Tp) of a first algorithm, wherein B (W) is the first heat of the comments, atp is the number of times that the comments are approved on the current day, btp is the number of times that the comments are approved on the past day, btp and Atp are summed to be the total number of the comments, tp is the ratio of the total of Btp and Atp of all the comments to the number thereof, and R is an average value of the ratio of Atp of all the comments to the total of Btp and Atp thereof; formula according to the second algorithm
Calculating a second degree of hotness of the comment, wherein
For the second popularity of the review, atp is the number of times the review was liked on the current day, btp is the number of times the review was liked on the past day; according to the formula of weighted averaging
Calculating a weighted popularity of the comment, wherein H (W) is the weighted popularity of the comment, α is a first weight, and β is a second weight; judging whether the comment is a hot comment or not according to the weighted popularity of the comment, and if the comment is a hot comment, passing the review of the recommendation record; calculating a first negative heat degree of the comment according to a formula B (W) = (Atp + Tp × R) ÷ ((Btp + Atp) + Tp) of a first algorithm, wherein B (W) is the first negative heat degree of the comment, atp is the number of times the comment is stepped on the day, btp is the number of times the comment is stepped on the past day, the total of Btp and Atp is the total number of times the comment is stepped on the past day, tp is the ratio of the total of Btp and Atp to the number of the comments, and R is an average value of the ratio of Atp of all comments to the total of Btp and Atp of the comments; judging whether the comment is a negative comment according to the first negative heat of the comment; if the review is a negative review, the review for the recommendation record fails.
2. An article generation device based on comment recommendation is characterized by comprising a foreground module and a background module; the foreground module carries out recommendation operation on the comments; inputting a recommended title of the comment; the background module submits a recommendation record of the comment; auditing the recommendation record based on the review; if the audit on the recommendation record is not passed, setting the state of the recommendation record as not passed; if the review of the recommended record is passed, entering the next step; entering the comments as a new article; adding a user selected topic to the new article; setting the state of the new article as pass; the foreground module displays the abstract of the new article in an article list;
the background module generates an identification number of the new article after setting the state of the new article as passing; recording the identification number of the new article into the recommendation record; the foreground module displays the comment in a comment recommending area; when the background module operates the comments, the background module acquires the identification numbers of the new articles from the recommendation records; the foreground module displays the full text of the new article;
after reviewing recommendation records based on comments, counting the number of times that the comments are approved on the current day and the past day, and calculating a first heat of the comments according to a formula B (W) = (Atp + Tp × R) ÷ ((Btp + Atp) + Tp) of a first algorithm, wherein B (W) is the first heat of the comments, atp is the number of times that the comments are approved on the current day, btp is the number of times that the comments are approved on the past day, btp and Atp are summed to be the total number of the comments, tp is the ratio of the total of Btp and Atp of all the comments to the number thereof, and R is an average value of the ratio of Atp of all the comments to the total of Btp and Atp thereof; formula according to the second algorithm
Calculating a second degree of hotness of the comment, wherein
For the second popularity of the review, atp is the number of times the review was liked on the current day, btp is the number of times the review was liked on the past day; according to the formula of weighted averaging
Calculating a weighted popularity of the comment, wherein H (W) is the weighted popularity of the comment, α is a first weight, and β is a second weight; judging whether the comment is a hot comment or not according to the weighted heat degree of the comment, and if the comment is the hot comment, passing the review of the recommendation record; calculating a first negative heat of the comment according to a formula B (W) = (Atp + Tp × R) ÷ ((Btp + Atp) + Tp) of a first algorithm, wherein B (W) is the first negative heat of the comment, atp is the number of times the comment is stepped on the day, btp is the number of times the comment is stepped on the past day, btp, atp sum is the total number of times the comment is stepped on the past day, tp is the ratio of the Btp, atp sum and the number thereof of all comments, and R is an average of the ratio of Atp and the Btp, atp sum thereof of all comments; judging whether the comment is a negative comment according to the first negative heat of the comment; if the comment is negativeAnd if the recommendation is reviewed, the review of the recommendation record is not passed.
3. The apparatus of claim 2, further comprising a database module; the database module stores the recommendation record; and the background module records the recommendation title information of the comments into the recommendation record.
4. A computer-readable storage medium, comprising a computer program which runs on a computer, the computer performing the method of claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110083047.9A CN112948706B (en) | 2021-01-21 | 2021-01-21 | Article generation method and device based on comment recommendation and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110083047.9A CN112948706B (en) | 2021-01-21 | 2021-01-21 | Article generation method and device based on comment recommendation and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112948706A CN112948706A (en) | 2021-06-11 |
| CN112948706B true CN112948706B (en) | 2023-01-31 |
Family
ID=76235789
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110083047.9A Active CN112948706B (en) | 2021-01-21 | 2021-01-21 | Article generation method and device based on comment recommendation and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112948706B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011142615A2 (en) * | 2010-05-14 | 2011-11-17 | Kim Ji Woong | Comment recommendation method using a virtual value index, system for the method, web server and system for providing a web service |
| CN104156390A (en) * | 2014-07-07 | 2014-11-19 | 乐视网信息技术（北京）股份有限公司 | Comment recommendation method and system |
| CN107491491A (en) * | 2017-07-20 | 2017-12-19 | 西南财经大学 | A kind of media article for adapting to user interest change recommends method |
| CN108153723A (en) * | 2017-12-27 | 2018-06-12 | 北京百度网讯科技有限公司 | Hot spot information comment generation method, device and terminal device |
| CN108920611A (en) * | 2018-06-28 | 2018-11-30 | 北京百度网讯科技有限公司 | article generation method, device, equipment and storage medium |
| CN109710841A (en) * | 2018-12-17 | 2019-05-03 | 北京百度网讯科技有限公司 | Comment on recommended method and device |
-
2021
- 2021-01-21 CN CN202110083047.9A patent/CN112948706B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011142615A2 (en) * | 2010-05-14 | 2011-11-17 | Kim Ji Woong | Comment recommendation method using a virtual value index, system for the method, web server and system for providing a web service |
| CN104156390A (en) * | 2014-07-07 | 2014-11-19 | 乐视网信息技术（北京）股份有限公司 | Comment recommendation method and system |
| CN107491491A (en) * | 2017-07-20 | 2017-12-19 | 西南财经大学 | A kind of media article for adapting to user interest change recommends method |
| CN108153723A (en) * | 2017-12-27 | 2018-06-12 | 北京百度网讯科技有限公司 | Hot spot information comment generation method, device and terminal device |
| CN108920611A (en) * | 2018-06-28 | 2018-11-30 | 北京百度网讯科技有限公司 | article generation method, device, equipment and storage medium |
| CN109710841A (en) * | 2018-12-17 | 2019-05-03 | 北京百度网讯科技有限公司 | Comment on recommended method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112948706A (en) | 2021-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9760547B1 (en) | Monetization of online content | |
| Shmueli et al. | Care to comment? Recommendations for commenting on news stories | |
| AU2008335085B2 (en) | User-created content aggregation and sharing | |
| US9323836B2 (en) | Internet based method and system for ranking artists using a popularity profile | |
| Mukherjee et al. | Effect of social networking advertisements on shaping consumers’ attitude | |
| US8321496B2 (en) | User evaluation of content on distributed communication network | |
| WO2009046434A1 (en) | Collaborative production of rich media content | |
| Currie et al. | Management education journals' rank and tier by active scholars | |
| US10134075B1 (en) | Method and system for exposing research note data based upon user activity data | |
| US20150302448A1 (en) | Sponsored content system and method for publication and interactive discussion engine | |
| US8688732B1 (en) | System for mining research data | |
| WO2015103419A1 (en) | Integrated media, publication and interactive discussion engine driven by user-specified topic | |
| Beaudouin et al. | Forms of contribution and contributors’ profiles: An automated textual analysis of amateur on line film critics | |
| Sattelberger | Optimising media marketing strategies in a multi-platform world: an inter-relational approach to pre-release social media communication and online searching | |
| CN112948706B (en) | Article generation method and device based on comment recommendation and storage medium | |
| Zhang et al. | The impact of virtual try-on image interaction technology on online shoppers' purchase decision | |
| US8140375B2 (en) | Voting on claims pertaining to a resource | |
| Verdaasdonk | Effects of acquired readership and reviewers' attention on the sales of new literary works | |
| US20150081687A1 (en) | System and method for user-generated similarity ratings | |
| Gurevych | OsCommerce Webmaster's Guide to Selling Online | |
| US20090187438A1 (en) | Method for review appraisals | |
| Bhatt | A study on customers’ perceptions towards E service quality dimensions and their satisfaction of online shopping platforms | |
| Mumay | Digital labor quality and news quality in Turkish digital mainstream media | |
| Calvert | Gfk Mediamark MRI+ University Internet Reporter | |
| Dinesha | Influence of Social Networks Recommendations for Consumers Purchase Decision in Retail Setting |
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 |