CN106777341A - Information processing method, device and computer equipment - Google Patents

Abstract

The invention discloses an information processing method, an information processing device and computer equipment. The information processing method comprises the following steps: obtaining user comments, traversing a comment queue, and judging whether the number of comments which are the same as or similar to the user comments in the comment queue reaches a first threshold value, wherein the comment queue is a first-in first-out queue and has a second threshold value in length, and if so, determining the user comments as spam comments; and if not, adding the user comment into a comment queue, and processing the comment at the tail of the first-in first-out queue according to the second threshold value. According to the embodiment of the invention, the spam comments can be effectively identified, and when the user comments are identified as non-spam comments, only the comment queue needs to be updated, so that all contents in the database are prevented from being processed, the operation load of the system is reduced, and the information processing efficiency is improved.

Description

Information processing method, device and computer equipment

technical field

The present invention relates to the field of communication technology, in particular to the field of Internet technology, and in particular to an information processing method, device and computer equipment.

Background technique

With the development of Internet technology, users can post relevant comments on various public platforms such as various forums, communities, and websites through the Internet. However, based on the openness of speech on the public platform, some users post a large number of spam comments such as advertising comments, promotional comments, and other comments with adverse effects on the public platform, thereby affecting users' acquisition of useful information and bringing users to adverse effects. During the use of existing computer equipment, spam comments have increasingly troubled users, and how to effectively identify spam comments has attracted more and more attention from the industry.

Contents of the invention

Embodiments of the present invention provide an information processing method, device, and computer equipment, which can improve information processing efficiency.

An embodiment of the present invention provides an information processing method, the method comprising:

Get user comments;

traversing the comment queue, and judging whether the number of comments in the comment queue that are the same as or similar to the user's comments reaches a first threshold, wherein the comment queue is a first-in-first-out queue and its length has a second threshold;

If so, then determining the user comments as spam comments;

If not, add the user comments into a comment queue, and process comments at the end of the first-in-first-out queue according to the second threshold.

An embodiment of the present invention also provides an information processing device, the device comprising:

Obtain module, used to obtain user comments;

The first judging module is used to traverse the comment queue and judge whether the number of comments in the comment queue that are the same as or similar to the user's comments reaches a first threshold, wherein the comment queue is a first-in-first-out queue and has a length of a second threshold ;

A determination module, configured to determine the user comments as spam comments when judging that the number of comments identical or similar to the user comments in the comment queue reaches a first threshold;

A processing module, configured to add the user comments to the comment queue when judging that the number of comments that are the same as or similar to the user comments in the comment queue does not reach the first threshold, and classify the advanced user comments according to the second threshold The comments at the end of the queue that go out of the queue first are processed.

An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored on the memory and operable on the processor, wherein the processor calls the computer program stored in the memory The program executes the information processing method described in any embodiment of the present invention.

Description of drawings

The technical solutions and other beneficial effects of the present invention will be apparent through the detailed description of specific embodiments of the present invention in conjunction with the accompanying drawings.

FIG. 1 is a schematic flowchart of an information processing method provided by an embodiment of the present invention.

Fig. 2 is a schematic diagram of a first usage state of an information processing method provided by an embodiment of the present invention.

Fig. 3 is a schematic diagram of a second usage state of an information processing method provided by an embodiment of the present invention.

FIG. 4 is another schematic flowchart of an information processing method provided by an embodiment of the present invention.

FIG. 5 is a schematic diagram of a third usage state of an information processing method provided by an embodiment of the present invention.

FIG. 6 is a schematic diagram of a fourth usage state of an information processing method provided by an embodiment of the present invention.

FIG. 7 is a schematic structural diagram of an information processing device provided by an embodiment of the present invention.

FIG. 8 is another schematic structural diagram of an information processing device provided by an embodiment of the present invention.

FIG. 9 is a schematic structural diagram of a computer device provided by an embodiment of the present invention.

detailed description

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. It should be understood that the specific embodiments described here are only used to explain the present invention, but not to limit the present invention. In addition, it should be noted that, for the convenience of description, only some structures related to the present invention are shown in the drawings but not all structures. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

The terms "first", "second" and "third" in the present invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "include" and "have", as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or modules is not limited to the listed steps or modules, but optionally also includes steps or modules that are not listed, or optionally includes For other steps or modules inherent in these processes, methods, products or devices.

Reference herein to an "embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present invention. The occurrences of this phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is understood explicitly and implicitly by those skilled in the art that the embodiments described herein can be combined with other embodiments.

The execution subject of an information processing method provided in an embodiment of the present invention may be an information processing device provided in an embodiment of the present invention, or a computer device integrating the information processing device (such as a desktop computer, a notebook, a palmtop computer, tablet computer, smart phone, etc.), the information processing device can be realized by hardware or software.

Please refer to FIG. 1 . FIG. 1 is a schematic flowchart of an information processing method provided by an embodiment of the present invention. The methods include:

Step S101, acquiring user comments.

Step S102, traversing the comment queue, and judging whether the number of comments in the comment queue that are the same as or similar to the user's comments reaches a first threshold, wherein the comment queue is a first-in-first-out queue and

The length has a second threshold; if not, execute step S103; if yes, execute step S104.

In some implementation manners, it may be determined whether the number of historical comments in the comment queue whose similarity with the user's comments reaches the third threshold reaches the first threshold, so as to determine whether the comments in the comment queue are similar to the user's comments. Whether the number of comments with the same or similar comments reaches the first threshold. When it is judged that the number of historical comments in the comment queue whose similarity with the user comments reaches the third threshold does not reach the first threshold, determine the same or similar comments in the comment queue as the user comments If the number does not reach the first threshold, step S103 is executed. When it is judged that the number of historical comments in the comment queue whose similarity with the user comments reaches the third threshold reaches the first threshold, determine the number of comments that are the same as or similar to the user comments in the comment queue When the first threshold is reached, step S104 is executed.

Step S103, adding the user comments into a comment queue, and processing comments at the end of the comment queue according to the second threshold.

In some implementation manners, the user comments may be added to the comment queue as the head comments, and the tail comments exceeding the second threshold are deleted.

Step S104, determining the user comments as spam comments.

In order to further understand the technical solution of the present invention, please refer to Figure 2 and Figure 3, Figure 2 is a schematic diagram of the first use state of an information processing method provided by an embodiment of the present invention, and Figure 3 is a kind of information provided by an embodiment of the present invention Schematic diagram of the second usage state of the processing method.

As shown in Figure 2, in a certain forum, "Flying Tiger" published an article titled "Sensors in Mobile Phones", and the user "Code Farmer" submitted an article on this forum as "Strive for another smog as soon as possible." Sensors" user reviews. The server of this forum traverses the comment queue, and judges whether the number of comments identical or similar to the user comments in the comment queue reaches the first threshold; if so, the user comments are determined to be spam comments; otherwise, the user comments are If it is determined that the comment is not spam, the user comment is added to the comment queue, and the comments at the end of the first-in-first-out queue are processed according to the second threshold.

For example, the comment queue is a first-in-first-out queue with a length of 1000 items.

As shown in Figure 3, when the user comment is determined to be a non-spam comment, the comment queue is updated, and the user comment "strive for another haze sensor as soon as possible" is added to the comment queue as displayed in the comment area The head of the team commented, and the comment at the end of the team with the earliest comment time and overflowing the 1000th comment "seeking formaldehyde sensor." was deleted.

In the embodiment of the present invention, by acquiring user comments and traversing the comment queue, it is judged whether the number of comments identical or similar to the user comments in the comment queue reaches the first threshold, wherein the comment queue is a first-in-first-out queue with a length of The second threshold, if so, then determine the user comments as spam comments; if not, add the user comments to the comment queue, and process the comments at the end of the first-in-first-out queue according to the second threshold . The embodiment of the present invention can effectively identify spam comments, and when user comments are identified as non-spam comments, only the comment queue needs to be updated, avoiding processing all content in the database, reducing system operation burden, and effectively improving information processing efficiency.

In some implementations, after the acquisition of user comments, it also includes:

Judging whether the comment information in the user comments exists in the blacklist database, and if so, determining the user comments as spam comments.

In some implementation manners, the judging whether the comment information in the user comments exists in the blacklist database includes:

Judging whether the user comments contain information matching the characteristic information in the blacklist database, and if so, determining the user comments as spam comments.

In some implementations, when the user comment is determined to be a spam comment, it is detected whether the user comment contains contact information, and if so, the contact information is added to the blacklist database as feature information.

In some embodiments, the comments similar to the user comments include historical comments whose similarity with the user comments reaches a third threshold.

All the above optional technical solutions may be combined in any way to form an optional embodiment of the present invention, which will not be repeated here.

Please refer to FIG. 4 . FIG. 4 is another schematic flowchart of an information processing method provided by an embodiment of the present invention. The methods include:

Step S201, acquiring user comments.

For example, as shown in Figure 2, in a certain forum, "Flying Tiger" published an article titled "Sensors in Mobile Phones", and the user "Code Farmer" submitted an article on this forum as "Strive for another one as soon as possible". Haze sensor" user comments, the forum's server obtains the user comments from the background.

Step S202, judging whether the comment information in the user comments exists in the blacklist database. If not, execute step S203; if yes, execute step S205.

It can be understood that the comment information in the user comments may include information such as user names, user IDs, comment content, and comment release time.

In some implementation manners, it is determined whether the user comments contain information matching the feature information in the blacklist database. If yes, execute step S205; if not, execute step S203.

At present, many public platforms support interactive behaviors between users, and the forms of said public platforms may include e-commerce platforms, forums, communities, websites, microblogs, post bars, blogs, application download platforms, and the like. For example, when a user registers on the website and passes the authentication, he has the user identity information of the website and becomes a user of the website. The user can display his user behavior on the website, such as publishing articles, publishing products, tweeting, posting, Reply to comments, etc., and you can also comment, like, etc. on other posted information. Regarding the above comment content, some users may post a large number of spam comments with the same or similar content, such as advertising comments, promotional comments, comments containing reactionary, violent, pornographic, hyperlinks, abuse, defamation and other adverse effects.

It can be understood that a blacklist library may be preset, and the blacklist library includes a plurality of feature information.

In some embodiments, the feature information includes any one or more of user name, user ID, contact information, keywords, and homonyms of keywords.

It can be understood that the format of the contact information may be a combination of letters and numbers, and the length exceeds 7 bytes. Such as phone number, mobile phone number, WeChat number, QQ number.

For example, the keywords may include hyperlinks and advertisement words, prohibited words, special symbols and the like.

For example, user comments submitted by users contain hyperlinks and advertising words, such as product promotion, store or website recommendation, company publicity, business promotion, etc. The hyperlink generally appears in the form of a URL, and there will be multiple consecutive English alphabet characters, such as http://..., and the "http" character is set as a keyword, which can be found by scanning the keywords in user comments. Detect whether it contains a hyperlink; if it contains a hyperlink, it means that the user comment may be a spam comment, and then further judge whether it contains an advertisement word. For the advertisement words, for example, QQ, special price, hot sale, Taobao, free shipping and other words are set as the keywords of the advertisement words, and the combination of any number and "yuan" is also set as the characteristic information. When the keyword is contained in the user comment, it is determined that the comment information in the user comment exists in the blacklist database, and step S205 is executed.

For example, the prohibited words are words containing personal attacks.

For example, when some users submit user comments, they may add special symbols in the middle of keywords or comments, so as to avoid the detection of spam comments by relevant platforms. Therefore, special symbols such as "★", "*", "#", "&" can be set as keywords and stored in the blacklist database as characteristic information.

For example, users may replace the original keywords with homonyms or near-sounds, so as to avoid the detection of spam comments by related platforms, such as "fishing master 3 逋鱼提线 Jiaweixin a5a7a9 class tiline". Therefore, for the above-mentioned situation containing homonyms or near-sounds, the homonyms of keywords can be set as feature information and stored in the blacklist database.

For example, the user comment submitted by the user on a certain forum is "invoicing on behalf of others, add Q (22222211)", and it is detected that the user comment contains information matching the contact information in the blacklist database, and then the steps are performed S205. For example, the user comment submitted by the user is "deep and good text, worth learning." If it is detected that the user comment does not contain information matching the characteristic information in the blacklist database, step S203 is performed.

As shown in Figure 2, in a forum, "Flying Tiger" published an article titled "Sensors in Mobile Phones", and user "Code Farmer" submitted an article on the forum titled "Strive for another fog as soon as possible." Haze Sensor" user review. When it is judged that the user comment submitted by the user does not contain information matching the feature information in the blacklist database, step S203 is executed.

Please refer to FIG. 5 , which is a schematic diagram of a third usage state of an information processing method provided by an embodiment of the present invention.

In a forum, "Flying Tiger" published an article titled "Sensors in Mobile Phones", and a user submitted an article on the forum titled "Fishing Master 3 Fish Lifting Line Jiawei Xin a5a7a9 Lesson Proposal" line" user reviews. When it is judged that the user comment submitted by the user contains information matching the characteristic information in the blacklist database, step S205 is executed.

In some embodiments, it is also possible to set a whitelist library to determine whether the comment information in the user comments exists in the whitelist library, if so, the user comments can be determined as non-spam comments; otherwise, the user comments can be Comment identified as spam.

For example, for user reviews of products, user reviews related to products are usually classified as useful information, so it is possible to check whether they are spam reviews by filtering associated words related to product descriptions, such as subject words or emotional words. For example, taking a product published on an e-commerce platform as an example, the subject words can be core nouns related to the product, and the subject terms about the standard description of the product can be stored in the whitelist database in advance. If it is detected that the user submits a If the review information of the user does not contain any subject words in the product standard description, the user review can be determined as spam; if it is detected that the review information submitted by the user for the product contains any one or more of the product standard description When subject words are used, the user comments can be determined to be non-spam comments.

For example, the emotional words include emotional words that the user really intends to express his subjective views, attitudes, feelings, emotions, and the like. For example, take the evaluation of products sold on a certain website as an example. The reviews of the products are people’s evaluations and discussions on product-related parameters and purchase experience. People can truly express their subjective views, attitudes, and feelings through reviews. , emotions, etc. Therefore, product reviews necessarily contain the emotions of the reviewers. The fewer emotional words, the more likely they belong to spam comments.

Step S203, traversing the comment queue, and judging whether the number of comments in the comment queue that are the same as or similar to the user's comment reaches a first threshold; if not, execute step S204; if yes, execute step S205.

It can be understood that the number of comments identical or similar to the user comments in the comment queue can be determined by detecting whether the comment queue contains historical comments that are the same or similar to the user comments. For example, when the comment information in the user comments does not exist in the blacklist database, there are still a large number of historical comments with the same or similar content as the user comments in the comment queue, and when the comment information with the same or similar content When the number of comments reaches a certain threshold, it will also hinder users from obtaining useful information. In fact, user comments with duplicate content can also be classified as spam comments. Therefore, in order to identify spam comments more accurately, it is possible to further detect whether there are historical comments identical or similar to the user comments in the comment queue, and determine whether the number of comments identical or similar to the user comments in the comment queue is reached the first threshold. Wherein, the comment queue is a first-in-first-out queue composed of historical comments.

In some implementations, the comments similar to user comments include historical comments whose similarity with the user comments reaches a third threshold. By judging whether the number of historical comments in the comment queue whose similarity with the user comments reaches the third threshold reaches the first threshold, it can be determined whether the comments in the comment queue are the same as or similar to the user comments. Whether the number of comments reaches the first threshold.

For example, the degree of similarity can be determined by comparing the degree of matching between user comments and information contained in historical comments in the comment queue. For example, the third threshold may be 80%. When the matching degree of the user comments and the information contained in the historical comments in the comment queue reaches 90%, it is determined to be similar; When the degree of matching of the included information reaches 100%, it is determined to be the same.

In some implementation manners, the comment queue may include any one of a chain queue and an array queue.

It can be understood that, in a programming language, a queue is a linear table, and the data elements of the queue are also called queue elements. Inserting a queue element into the queue is called enqueue, and removing a queue element from the queue is called dequeue. Because the queue can only be inserted at one end and deleted at the other end, that is, the elements that enter the queue at the earliest can be deleted from the queue first, so the queue is also called FIFO—first in first out (FIFO) linear table. Therefore, the comment queue can be called a first-in first-out queue.

For example, a queue can be stored in an array Q[1...m], and the upper bound m of the array is the maximum capacity allowed by the queue. In the operation of the queue, two pointers need to be set: head, the pointer of the head of the queue, which points to the actual head element of the queue; tail, the pointer of the tail of the queue, which points to the next position of the actual tail element of the queue. Under normal circumstances, the initial value of the two pointers is set to 0, and the queue is empty at this time, with no elements. When the number of queue elements reaches the upper bound m of the array, when a new queue element enters the queue, the queue element that first enters the queue is deleted from the queue.

For example, the queue can also be stored in a linked list, and the mathematically logical sequential and adjacent relationship of the data is indicated by the pointer of the storage address of the element, thereby forming a chained queue, which can dynamically allocate storage.

For example, if the comment queue is an array queue, the second threshold of the length of the comment queue is the maximum capacity of the array queue, for example, 1000 user comments.

For example, when it is detected that the comment queue contains the same historical comments as the user comments, in order to avoid multiple user comments with repeated content in the comment queue, thereby affecting the user's information acquisition efficiency, it is possible to refuse to update the comment queue, And add 1 to the like array that records the same historical comments as the user comments, to indicate that other people have published user comments that are the same or similar to the content of the historical comments, or to indicate that other people agree with the historical comments Content.

As shown in Figure 2, such as the first threshold value is 5, and the number of comments identical to the user comments whose content is "strive for another haze sensor as soon as possible" in the comment queue is 1, then it is determined that the comments in the comment queue are the same as the user comments. If the number of comments with the same or similar comments from the user does not reach the first threshold, step S204 is performed.

As shown in Figure 5, for example, the first threshold is 5, and the number of comments in the comment queue that is the same as the user comment whose content is "fishing expert 3 fish lifting line Jiawei new a5a7a9 class lifting line" is 7 If it is determined that the number of comments identical or similar to the user's comments in the comment queue has reached a first threshold, step S205 is executed.

Step S204, adding the user comment into the comment queue as the head comment, and deleting the tail comment exceeding the second threshold.

It can be understood that the length of the first-in-first-out queue may be preset as the second threshold. The length can be represented by the total number of data packets that can be accommodated in the array queue. Before the array is established, it needs to be set to a fixed size in advance, that is, an appropriate byte length is set for each queue element to meet the requirements of a single queue element. The requirement for byte length can be understood as each queue element represents a data packet, and each data packet has a fixed size, for example, if the array is N[1...1000], then the second threshold is 1000. The length of the first-in-first-out queue can also be represented by the number of pointers of storage units in the chained queue. The linked list does not need to allocate a fixed-size storage space in advance. When data needs to be stored, a queue element can be set for each queue element. Appropriate storage units are used to store data, and the storage units are linked together with other storage units in the queue through pointers. The content of the comment queue changes in real time. For example, when a new user comment enters the queue in the area displayed in the comment area, the user comment is added to the comment queue as the team head comment and as the history of the team tail comment Comments are queued, and the queue numbers of other historical comments are increased by 1 on the basis of the original ones.

As shown in Figure 3, when the number of comments identical or similar to the user comments in the comment queue is less than the first threshold, the comment queue is updated, and the user comment "strive for another haze sensor as soon as possible" Add to the head No.1 of the comment queue, and delete the historical comment "seeking formaldehyde sensor" located at the tail No.1000 of the comment queue. The number of the historical comment "Good article! Like" originally numbered No.1 is changed to No.2, which is displayed in the display column of No.2, and the other historical comments are all moved backward by one display column.

Step S205, determining the user comments as spam comments.

It can be understood that, when it is determined that the user comments are spam comments, updating the comment queue may be refused.

Please refer to FIG. 6 . FIG. 6 is a schematic diagram of a fourth usage state of an information processing method provided by an embodiment of the present invention.

When it is determined that the user comment with the content of "fishing master 3 fish tiling line Jiawei new a5a7a9 class tiling line" is a spam comment, it refuses to update the comment queue.

In some implementation manners, when the update of the comment queue is refused, a prompt box may also pop up to remind the user of the failure to publish the comment information. As shown in Figure 6, when the user clicks the "Comment" button, a prompt box pops up with the content "Comment Review Failed: Spam Comment!", and refuses to update the comment queue at the same time. The comments section of .

Step S206, detecting whether the user comments contain contact information, and if so, adding the contact information to the blacklist database as feature information.

In some embodiments, when it is detected that the user comments contain contact information, and the contact information is a new contact information, the new contact information extracted from the user comments is added to the blacklist. List database as feature information. When the contact information is an old contact information, the original contact information in the blacklist database may be overwritten, or not added to the blacklist database.

It can be understood that, when a new contact method is detected in the user comment, the new contact method is extracted and added to the blacklist database as feature information to be used as the detection basis for the next user comment .

As shown in Figure 6, for example, when the user comment with the content of "fishing expert 3 fish tiling line Jiawei new a5a7a9 class tiling line" is a spam comment, the new contact information "a5a7a9" in the user comment is extracted, And "a5a7a9" is added to the blacklist library as feature information.

The embodiment of the present invention determines whether the user comment is a spam comment by detecting whether the user comment contains information that matches the characteristic information in the blacklist database, and traverses the comment queue when the user comment is a non-spam comment , and when it is judged that the number of comments that are the same as or similar to the user’s comments in the comment queue does not reach the first threshold, the user’s comment will be added to the comment queue as the leader comment, and will overflow the second threshold Comments at the end of the queue will be deleted. In the embodiment of the present invention, when the user comments submitted by the user are identified as non-spam comments, historical comments with duplicate content are further detected, so as to eliminate user comments that avoid adding a large amount of duplicate content, reduce the burden on system operation, effectively improve information processing efficiency, and improve The efficiency with which users obtain useful information.

An embodiment of the present invention also provides an information processing device, as shown in FIG. 7 , which is a schematic structural diagram of an information processing device provided by an embodiment of the present invention. The information processing device 30 includes an acquisition module 31 , a first judgment module 33 , a processing module 34 , and a determination module 35 .

Wherein, the acquiring module 31 is configured to acquire user comments.

The first judging module 33 is configured to traverse the comment queue, and judge whether the number of comments identical or similar to the user comments in the comment queue reaches a first threshold, wherein the comment queue is a first-in-first-out queue with a length of second threshold.

The determination module 35 is configured to determine the user comment as a spam comment when it is judged that the number of comments identical or similar to the user comment in the comment queue reaches a first threshold.

The processing module 34 is configured to add the user comments to the comment queue when it is judged that the number of comments that are the same as or similar to the user comments in the comment queue does not reach the first threshold, and add the user comments to the comment queue according to the second threshold. Comments at the end of the comment queue are processed.

Please refer to FIG. 8 . FIG. 8 is another schematic structural diagram of an information processing device provided by an embodiment of the present invention. The information processing device 30 includes an acquisition module 31 , a second judgment module 32 , a first judgment module 33 , a processing module 34 , a determination module 35 , and a detection module 36 .

Wherein the acquisition module 31 is used to acquire user comments.

For example, as shown in Figure 2, in a certain forum, "Flying Tiger" published an article titled "Sensors in Mobile Phones", and the user "Code Farmer" submitted an article on this forum as "Strive for another one as soon as possible". haze sensor", the acquisition module 31 acquires the user comments.

The second judging module 32 is configured to judge whether the comment information in the user comments exists in the blacklist database, and if so, determine the user comments as spam comments.

It can be understood that the comment information in the user comments may include information such as user names, user IDs, comment content, and comment release time.

In some implementations, the second judging module 32 is further configured to judge whether the user comments contain information that matches the characteristic information in the blacklist database, and if so, determine the user comments as spam comments.

At present, many public platforms support interactive behaviors between users, and the forms of said public platforms may include e-commerce platforms, forums, communities, websites, microblogs, post bars, blogs, application download platforms, and the like. For example, when a user registers on the website and passes the authentication, he has the user identity information of the website and becomes a user of the website. The user can display his user behavior on the website, such as publishing articles, publishing products, tweeting, posting, Reply to comments, etc., and you can also comment, like, etc. on other posted information. Regarding the above comment content, some users may post a large number of spam comments with the same or similar content, such as advertising comments, promotional comments, comments containing reactionary, violent, pornographic, hyperlinks, abuse, defamation and other adverse effects.

It can be understood that a blacklist library may be preset, and the blacklist library includes a plurality of feature information.

In some embodiments, the feature information includes any one or more of user name, user ID, contact information, keywords, and homonyms of keywords.

It can be understood that the format of the contact information may be a combination of letters and numbers, and the length exceeds 7 bytes. Such as phone number, mobile phone number, WeChat number, QQ number.

For example, the keywords may include hyperlinks and advertisement words, prohibited words, special symbols and the like.

For example, user comments submitted by users contain hyperlinks and advertising words, such as product promotion, store or website recommendation, company publicity, business promotion, etc. The hyperlink generally appears in the form of a URL, and there will be multiple consecutive English alphabet characters, such as http://..., and the "http" character is set as a keyword, which can be found by scanning the keywords in user comments. Detect whether it contains a hyperlink; if it contains a hyperlink, it means that the user comment may be a spam comment, and then further judge whether it contains an advertisement word. For the advertisement words, for example, QQ, special price, hot sale, Taobao, free shipping and other words are set as the keywords of the advertisement words, and the combination of any number and "yuan" is also set as the characteristic information. When the keyword is included in the user comment, the second judging module 32 determines that the comment information in the user comment exists in the blacklist database, and then determines the user comment as a spam comment.

For example, the prohibited words are words containing personal attacks.

For example, when some users submit user comments, they may add special symbols in the middle of keywords or comments, so as to avoid the detection of spam comments by relevant platforms. Therefore, special symbols such as "★", "*", "#", "&" can be set as keywords and stored in the blacklist database as characteristic information.

For example, users may replace the original keywords with homonyms or near-sounds, so as to avoid the detection of spam comments by related platforms, such as "fishing master 3 逋鱼提线 Jiaweixin a5a7a9 class tiline". Therefore, for the above-mentioned situation containing homonyms or near-sounds, the homonyms of keywords can be set as feature information and stored in the blacklist database.

For example, the user comment submitted by the user on a certain forum is "invoicing on behalf of others, add Q (22222211)", when the second judging module 32 determines that the user comment contains information related to the contact information in the blacklist database, If there is no matching information, the user comment is determined to be a spam comment.

As shown in Figure 5, in a certain forum, "Flying Tiger" published an article titled "Sensors in Mobile Phones", and the user submitted an article on this forum with the content "Fishing Master 3 Fish Lifting Line" Jiawei's new a5a7a9 class lift line" user reviews. When the second judging module 32 judges that the user comment submitted by the user contains information that matches the feature information in the blacklist database, the user comment is determined as a spam comment.

In some implementation manners, a whitelist library may also be set. The second judging module 32 can also be used to judge whether the comment information in the user comments exists in the whitelist library, if so, the user comments can be determined as non-spam comments; otherwise, the user comments can be Confirmed as spam.

For example, for user reviews of products, user reviews related to products are usually classified as useful information, so it is possible to check whether they are spam reviews by filtering associated words related to product descriptions, such as subject words or emotional words. For example, taking a product published on an e-commerce platform as an example, the subject words can be core nouns related to the product, and the subject terms about the standard description of the product can be stored in the whitelist database in advance. If it is detected that the user submits a If the review information of the user does not contain any subject words in the product standard description, the user review can be determined as spam; if it is detected that the review information submitted by the user for the product contains any one or more of the product standard description When subject words are used, the user comments can be determined to be non-spam comments.

For example, the emotional words include emotional words that the user really intends to express his subjective views, attitudes, feelings, emotions, and the like. For example, take the evaluation of products sold on a certain website as an example. The reviews of the products are people’s evaluations and discussions on product-related parameters and purchase experience. People can truly express their subjective views, attitudes, and feelings through reviews. , emotions, etc. Therefore, product reviews necessarily contain the emotions of the reviewers. The fewer emotional words, the more likely they belong to spam comments.

The first judging module 33 is configured to traverse the comment queue, and judge whether the number of comments identical or similar to the user comments in the comment queue reaches a first threshold, wherein the comment queue is a first-in-first-out queue with a length of second threshold.

It can be understood that the number of comments identical or similar to the user comments in the comment queue can be determined by detecting whether the comment queue contains historical comments that are the same or similar to the user comments. For example, when the comment information in the user comments does not exist in the blacklist database, there are still a large number of historical comments with the same or similar content as the user comments in the comment queue, and when the comment information with the same or similar content When the number of comments reaches a certain threshold, it will also hinder users from obtaining useful information. In fact, user comments with duplicate content can also be classified as spam comments. Therefore, in order to identify spam comments more accurately, it is possible to further detect whether there are historical comments identical or similar to the user's comments in the comment queue, and judge whether the comment queue is similar to the user's comments in the comment queue by the first judging module 33. Whether the number of comments with the same or similar comments reaches the first threshold. Wherein, the comment queue is a first-in-first-out queue composed of historical comments.

In some implementations, the comments similar to user comments include historical comments whose similarity with the user comments reaches a third threshold. It can be determined by the first judging module 33 whether the number of historical comments in the comment queue whose similarity with the user comments reaches the third threshold reaches the first threshold, so as to determine whether the comments in the comment queue are similar to the user comments. Whether the number of comments with the same or similar comments of the above user reaches the first threshold.

For example, the degree of similarity can be determined by comparing the degree of matching between user comments and information contained in historical comments in the comment queue. For example, the third threshold may be 80%. When the matching degree of the user comments and the information contained in the historical comments in the comment queue reaches 90%, it is determined to be similar; When the degree of matching of the included information reaches 100%, it is determined to be the same.

In some implementation manners, the comment queue may include any one of a chain queue and an array queue.

It can be understood that, in a programming language, a queue is a linear table, and the data elements of the queue are also called queue elements. Inserting a queue element into the queue is called enqueue, and removing a queue element from the queue is called dequeue. Because the queue can only be inserted at one end and deleted at the other end, that is, the elements that enter the queue at the earliest can be deleted from the queue first, so the queue is also called FIFO—first in first out (FIFO) linear table. Therefore, the comment queue can be called a first-in first-out queue.

For example, a queue can be stored in an array Q[1...m], and the upper bound m of the array is the maximum capacity allowed by the queue. In the operation of the queue, two pointers need to be set: head, the pointer of the head of the queue, which points to the actual head element of the queue; tail, the pointer of the tail of the queue, which points to the next position of the actual tail element of the queue. Under normal circumstances, the initial value of the two pointers is set to 0, and the queue is empty at this time, with no elements. When the number of queue elements reaches the upper bound m of the array, when a new queue element enters the queue, the queue element that first enters the queue is deleted from the queue.

For example, the queue can also be stored in a linked list, and the mathematically logical sequential and adjacent relationship of the data is indicated by the pointer of the storage address of the element, thereby forming a chained queue, which can dynamically allocate storage.

For example, if the comment queue is an array queue, the second threshold of the length of the comment queue is the maximum capacity of the array queue, for example, 1000 user comments.

For example, when it is detected that the comment queue contains the same historical comments as the user comments, in order to avoid multiple user comments with repeated content in the comment queue, thereby affecting the user's information acquisition efficiency, it is possible to refuse to update the comment queue, And add 1 to the like array that records the same historical comments as the user comments, to indicate that other people have published user comments that are the same or similar to the content of the historical comments, or to indicate that other people agree with the historical comments Content.

As shown in Figure 2, for example, the first threshold value is 5, and the number of comments identical to the user comments whose content is "strive for another haze sensor as soon as possible" in the comment queue is 1, then the first judging module 33 It is determined that the number of comments identical or similar to the user's comments in the comment queue does not reach the first threshold.

As shown in Figure 5, for example, the first threshold is 5, and the number of comments in the comment queue that is the same as the user comment whose content is "fishing expert 3 fish lifting line Jiawei new a5a7a9 class lifting line" is 7 , the first judging module 33 judges that the number of comments that are the same as or similar to the user's comments in the comment queue has reached a first threshold.

The processing module 34 is configured to add the user comments into the comment queue as the head comments, and delete the tail comments exceeding the second threshold.

It can be understood that the length of the FIFO queue may be preset as the second threshold. The length can be represented by the total number of data packets that can be accommodated in the array queue. Before the array is established, it needs to be set to a fixed size in advance, that is, an appropriate byte length is set for each queue element to meet the requirements of a single queue element. The requirement for byte length can be understood as each queue element represents a data packet, and each data packet has a fixed size, for example, if the array is N[1...1000], then the second threshold is 1000. The length of the first-in-first-out queue can also be represented by the number of pointers to the storage units in the chained queue. The linked list does not need to allocate a fixed-size storage space in advance. When data needs to be stored, one can be set for each queue element. Appropriate storage units are used to store data, and the storage units are linked together with other storage units in the queue through pointers. The content of the comment queue changes in real time. For example, when a new user comment enters the queue in the area displayed in the comment area, the processing module 34 adds the user comment to the comment queue as the leader comment, as The historical comments of the comments at the end of the queue will be dequeued, and the queue numbers of other historical comments will be increased by 1 on the original basis.

As shown in FIG. 3 , when the number of comments that are the same as or similar to the user comments in the comment queue is less than the first threshold, the comment queue is updated, and the processing module 34 puts the user comment "try to come back sooner." Haze sensor" is added to the head No.1 of the comment queue, and the historical comment "seeking formaldehyde sensor" located at the tail No.1000 of the comment queue is deleted. The number of the historical comment "Good article! Like" originally numbered No.1 is changed to No.2, which is displayed in the display column of No.2, and the other historical comments are all moved backward by one display column.

The determination module 35 is configured to determine the user comment as a spam comment when it is judged that the number of comments identical or similar to the user comment in the comment queue reaches a first threshold.

It can be understood that, when the determining module 35 determines that the user comment is spam, the processing module 34 may refuse to update the comment queue.

As shown in FIG. 6 , when the determination module 35 determines that the user comment whose content is "fishing master 3 耋鱼提线 Jiaweixin a5a7a9 class tiline" is a spam comment, the processing module 34 refuses to comment The comment queue is updated.

In some implementations, when the processing module 34 refuses to update the comment queue, a prompt box may also pop up to remind the user of the failure to publish the comment information. As shown in Figure 6, when the user clicks the "Comment" button, a prompt box pops up with the content "Comment Review Failed: Spam Comment!", and refuses to update the comment queue at the same time. The comments section of .

The detection module 36 is configured to detect whether the user comment contains contact information when the user comment is determined to be a spam comment, and if so, add the contact information to a blacklist database as characteristic information.

In some implementations, when the detection module 36 detects that the user comments contain contact information, and the contact information is a new contact information, the new contact information extracted from the user comments is updated to Added to the blacklist library as feature information. When the contact information is an old contact information, the original contact information in the blacklist database may be overwritten, or not added to the blacklist database.

It can be understood that, when a new contact method is detected in the user comment, the new contact method is extracted and added to the blacklist database as feature information to be used as the detection basis for the next user comment .

As shown in Figure 6, for example, when the user comment with the content of "fishing expert 3 fish tiling line Jiawei new a5a7a9 class tiling line" is a spam comment, the new contact information "a5a7a9" in the user comment is extracted, And "a5a7a9" is added to the blacklist library as feature information.

An embodiment of the present invention also provides a computer device, as shown in FIG. 9 , which is a schematic structural diagram of a computer device provided by an embodiment of the present invention. The computer device 400 may include a radio frequency (RF, Radio Frequency) circuit 401, a memory 402 including one or more computer-readable storage media, an input unit 403, a display unit 404, a sensor 405, an audio circuit 406, a wireless fidelity ( WiFi (Wireless Fidelity) module 407, including a processor 408 with one or more processing cores, and a power supply 409 and other components. Those skilled in the art can understand that the structure of the computer device shown in FIG. 9 is not limited to the computer device, and may include more or less components than shown in the figure, or combine certain components, or arrange different components.

The radio frequency circuit 401 can be used for sending and receiving information, or receiving and sending signals during a call.

Memory 402 may be used to store applications and data. The application programs stored in the memory 402 include computer programs.

The input unit 403 can be used to receive input numbers, character information or user characteristic information (such as fingerprints), and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

The display unit 404 can be used to display information input by or provided to the user and various graphical user interfaces of the computer device. These graphical user interfaces can be composed of graphics, text, icons, videos and any combination thereof.

The computer device may also include at least one sensor 405, such as a light sensor, motion sensor, and other sensors.

The audio circuit 406 may provide an audio interface between the user and the computer device through speakers, microphones.

The wireless fidelity (WiFi) module 407 can be used for short-distance wireless transmission, and can help users send and receive emails, browse websites and access streaming media, etc., and it provides users with wireless broadband Internet access.

The processor 408 is the control center of the computer equipment. It uses various interfaces and lines to link various parts of the entire computer equipment. By running or executing the application programs stored in the memory 402 and calling the data stored in the memory 402, the computer equipment executes Various functions and processing data, so as to monitor the computer equipment as a whole.

The computer device also includes a power source 409 (such as a battery) for powering various components.

Although not shown in FIG. 9 , the computer device may also include a camera, a Bluetooth module, etc., which will not be repeated here.

Specifically, in this embodiment, the processor 408 in the computer device loads the computer program corresponding to the process of one or more application programs into the memory 402 according to the following instructions, and the processor 408 runs the computer program stored in the memory. 402 in the application, perform the following operations:

Get user comments;

traversing the comment queue, and judging whether the number of comments in the comment queue that are the same as or similar to the user's comments reaches a first threshold, wherein the comment queue is a first-in-first-out queue and its length has a second threshold;

If so, then determining the user comments as spam comments;

If not, add the user comment into a comment queue, and process comments at the end of the comment queue according to the second threshold.

In some implementations, the processor 408 is configured to add the user comments to the comment queue, and process comments at the end of the comment queue according to the second threshold, including:

Adding the user's comment into the comment queue as the head comment, and deleting the tail comment exceeding the second threshold.

In some implementations, the processor 408 is configured to further include:

Judging whether the comment information in the user comments exists in the blacklist database, and if so, determining the user comments as spam comments.

In some implementations, the processor 408 is used to determine whether the comment information in the user comments exists in the blacklist library, including:

Judging whether the user comments contain information matching the feature information in the blacklist database, and if so, determining that the comment information in the user comments exists in the blacklist database.

In some implementations, the processor 408 is also used to:

When the user comment is determined to be a spam comment, it is detected whether the user comment contains contact information, and if so, the contact information is added to the blacklist library.

In some implementations, the processor 408 is used to determine whether the number of comments that are the same as or similar to the user's comments in the comment queue reaches a first threshold, including:

Judging whether the number of historical comments whose similarity with the user comments in the comment queue reaches the third threshold reaches the first threshold, and if so, determine the comments that are the same as or similar to the user comments in the comment queue The number reaches the first threshold.

In the foregoing embodiments, the descriptions of each embodiment have their own emphases, and for parts not described in detail in a certain embodiment, reference may be made to relevant descriptions of other embodiments.

In the embodiment of the present invention, the information processing device and the information processing method in the above embodiments belong to the same idea, any method provided in the information processing method embodiment can be run on the information processing device, For the specific implementation process, refer to the embodiment of the information processing method, and details are not repeated here.

It should be noted that, for the information processing method of the present invention, ordinary testers in the field can understand that all or part of the flow of the information processing method described in the embodiment of the present invention can be completed by controlling the relevant hardware through a computer program , the computer program may be stored in a computer-readable storage medium, such as stored in a memory of a computer device, and executed by at least one processor in the computer device, and the execution process may include information processing as described Flow of an embodiment of the method. Wherein, the storage medium may be a magnetic disk, an optical disk, a read only memory (ROM, Read Only Memory), a random access memory (RAM, Random Access Memory) and the like.

For the information processing device in the embodiment of the present invention, its various functional modules may be integrated into one processing chip, or each module may exist separately physically, or two or more modules may be integrated into one module. The above-mentioned integrated modules can be implemented in the form of hardware or in the form of software function modules. If the integrated modules are implemented in the form of software function modules and sold or used as independent products, they can also be stored in a computer-readable storage medium, such as read-only memory, magnetic disk or optical disk, etc. .

The information processing method, device and computer equipment provided by the embodiment of the present invention have been introduced in detail above. In this paper, specific examples are used to illustrate the principle and implementation of the present invention. The description of the above embodiment is only for helping Understand the technical solution and its core idea of the present invention; those skilled in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or The replacement does not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.

Claims (13)

1. a kind of information processing method, it is characterised in that methods described includes：

Obtain user comment；

Traversal comment queue, judge it is described comment on queue in the number of reviews same or similar with the user comment whether reach To first threshold, wherein the comment queue is fifo queue and length has Second Threshold；

If so, the user comment then is defined as into comment spam；

If it is not, the user comment is then added into comment queue, and according to the Second Threshold to the tail of the queue for commenting on queue Comment is processed.

2. information processing method according to claim 1, it is characterised in that described that the user comment is added into comment team Row, and the tail of the queue comment for commenting on queue is processed according to the Second Threshold, including：

The user comment is added the comment queue commented on as head of the queue, and the tail of the queue comment of the Second Threshold will be overflowed Deleted.

3. information processing method according to claim 1, it is characterised in that after the acquisition user comment, also wrap Include：

The comment information in the user comment is judged with the presence or absence of in blacklist storehouse, if being then defined as the user comment Comment spam.

4. information processing method according to claim 1, it is characterised in that the comment in the judgement user comment Information whether there is in blacklist storehouse, including：

Judge whether include the information matched with the characteristic information in blacklist storehouse in the user comment, if then by institute State user comment and be defined as comment spam.

5. the information processing method according to claim any one of 1-4, it is characterised in that methods described also includes：

When the user comment is defined as comment spam, whether contain contact method in the detection user comment, if then The contact method is added in blacklist storehouse as characteristic information.

6. information processing method according to claim 1, it is characterised in that the comment similar to user comment includes With the historical review that the similarity of the user comment reaches the 3rd threshold value.

7. a kind of information processor, it is characterised in that described device includes：

Acquisition module, for obtaining user comment；

First judge module, for traveling through comment queue, judges or phase identical with the user comment in the comment queue As number of reviews whether reach first threshold, wherein the comment queue is that fifo queue and length have the second threshold Value；

Determining module, reaches for same or similar with user comment number of reviews in the comment queue is judged During first threshold, the user comment is defined as comment spam；

Processing module, does not reach for same or similar with user comment number of reviews in the comment queue is judged During to first threshold, the user comment is added into comment queue, and according to the Second Threshold to the team for commenting on queue Tail comment is processed.

8. information processor according to claim 7, it is characterised in that the processing module, for by the user Comment adds the comment queue to be commented on as head of the queue, and will overflow the tail of the queue comment of the Second Threshold and be deleted.

9. information processor according to claim 7, it is characterised in that described device also includes：

Second judge module, for judging the comment information in the user comment with the presence or absence of in blacklist storehouse, if then will The user comment is defined as comment spam.

10. information processor according to claim 7, it is characterised in that second judge module, for judging State and whether include the information matched with the characteristic information in blacklist storehouse in user comment, if then by the user comment It is defined as comment spam.

11. information processor according to claim any one of 7-10, it is characterised in that described device also includes：

Detection module, for when the user comment is defined as comment spam, whether connection being contained in the detection user comment It is mode, if then the contact method is added in blacklist storehouse as characteristic information.

12. information processors according to claim 7, it is characterised in that the comment bag similar to user comment Include the historical review that the 3rd threshold value is reached with the similarity of the user comment.

A kind of 13. computer equipments, including memory, processor and storage can run on a memory and on the processor Computer program, it is characterised in that the processor calls the computer program stored in the memory, performs such as Information processing method described in claim any one of 1-6.