CN110516156A - A kind of network behavior monitoring device, method, equipment and storage medium - Google Patents
A kind of network behavior monitoring device, method, equipment and storage medium Download PDFInfo
- Publication number
- CN110516156A CN110516156A CN201910810292.8A CN201910810292A CN110516156A CN 110516156 A CN110516156 A CN 110516156A CN 201910810292 A CN201910810292 A CN 201910810292A CN 110516156 A CN110516156 A CN 110516156A
- Authority
- CN
- China
- Prior art keywords
- sensitive
- user
- log information
- behaviors log
- information
- 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.)
- Granted
Links
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
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- 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
- G06F16/955—Retrieval from the web using information identifiers, e.g. uniform resource locators [URL]
- G06F16/9566—URL specific, e.g. using aliases, detecting broken or misspelled links
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 embodiment of the invention discloses a kind of network behavior monitoring device, method, equipment and computer readable storage medium, the user behaviors log information that will acquire is matched with preset sensitive keys words group, determines sensitive users；User behaviors log information is generated by the network behavior of user.According to the property parameters of all kinds of label informations, the value-at-risk of each sensitive users is determined；Label information divides to obtain according to the information attribute that the historical behavior log information of each user includes.Value-at-risk is higher, illustrate sensitive users make abnormal behaviour probability it is bigger.In order to reduce the generation of sensitive users fortuitous event, value-at-risk can be monitored more than the user behaviors log information of the sensitive users of threshold value, when there is user behaviors log information matched with sensitive keys words group again in the sensitive users that value-at-risk is more than threshold value, then carry out warning note, to notification event, processing people quickly intervenes solution, effectively avoids the generation of fortuitous event.
Description
Technical field
The present invention relates to technical field of data processing, more particularly to a kind of network behavior monitoring device, method, equipment and
Computer readable storage medium.
Background technique
Social information of today is explosion, and various forms of stimulations emerge one after another, can all cause stress to sensitive group.
Sensitive group often makes abnormal behaviour under great psychological pressure, damages to itself, or even brings to society
Undesirable influence.
In the information age, internet becomes the important tool of people's communication, and many behaviors of sensitive group all exist
It is embodied in internet access, for example there is the sensitive group for tendency of committing suicide, often outwardly issued before committing suicide each
Kind help information, and these information are embodied often through network, for example are chatted, posted.If can find in time quick
Touching member can effectively avoid unnecessary unexpected generation to help sensitive personnel in time.
It is those skilled in the art's problem to be solved as it can be seen that how to find the abnormal behaviour of sensitive group in time.
Summary of the invention
The purpose of the embodiment of the present invention is that providing a kind of network behavior monitoring device, method, equipment and computer-readable depositing
Storage media can find the abnormal behaviour of sensitive group in time.
In order to solve the above technical problems, the embodiment of the present invention provides a kind of network behavior monitoring device, including matching unit,
Determination unit, tracing unit and Alarm Unit；
The matching unit, user behaviors log information and the progress of preset sensitive keys words group for will acquire
Match, determines sensitive users；Wherein, the user behaviors log information is generated by the network behavior of user；
The determination unit determines the value-at-risk of each sensitive users for the property parameters according to all kinds of label informations；
Wherein, the label information divides to obtain according to the information attribute that the historical behavior log information of each user includes；
The tracing unit is monitored for the user behaviors log information to value-at-risk more than the sensitive users of threshold value；
The Alarm Unit, for carrying out when there is user behaviors log information matched with the sensitive keys words group
Warning note.
Optionally, the matching unit includes filtering subelement, participle subelement and determines subelement；
The filtering subelement, it is corresponding with designated domain name for filtering out specified application from the user behaviors log information of acquisition
User behaviors log information；
The participle subelement obtains multiple participle groups for carrying out word segmentation processing to filtered user behaviors log information；
Wherein, each user has its corresponding participle group；
The determining subelement is determined for matching each participle group with preset sensitive keys words group
Sensitive users out.
Optionally, the determining subelement is for counting the sensitivity to match in target participle group with sensitive keys words group
Segment the number occurred；Wherein, target participle group is any one participle group in all participle groups；
When the number that sensitive participle occurs is greater than or equal to preset threshold, then the corresponding use of the target participle group is determined
Family is sensitive users；
When the number that sensitive participle occurs is less than the preset threshold, then judge be in target histories user behaviors log information
It is no the sensitive participle occur；Wherein, the target histories user behaviors log information is the corresponding user's of the target participle group
The user behaviors log information that web-based history behavior generates；
When occurring the sensitive participle in target histories user behaviors log information, then the corresponding use of the target participle group is determined
Family is sensitive users.
Optionally, the determining subelement is for counting the sensitivity to match in target participle group with sensitive keys words group
The total number of participle；Wherein, target participle group is any one participle group in all participle groups；
When the total number of sensitivity participle is greater than pre-set limit, then determine that the corresponding user of the target participle group is sensitivity
User.
Optionally, it is directed to the determination process of the property parameters of all kinds of label informations, described device further includes obtaining
Unit, training unit and as unit；
The acquiring unit, for obtaining sample data；Wherein, the sample data includes the row for being provided with sensitivity label
For the user behaviors log information of log information and not set sensitivity label；
The training unit, for being instructed using the sample data to the detection model for being provided with all kinds of label informations
Practice, until the accuracy rate of the detection model meets preset requirement, then triggers described as unit；
It is described to be used as unit, it is used for the corresponding attribute score of label informations all kinds of in trained detection model and weight
Value is used as property parameters.
Optionally, the determination unit includes coupling subelement and summation subelement；
The coupling subelement, for the user behaviors log information of each sensitive users to be matched with all kinds of label informations,
Determine every attribute score of each sensitive users；Wherein, every class label packet contains at least one attribute information, each category
Property information has its corresponding attribute score；And every class label information has its corresponding weighted value；
The summation subelement, for adding every attribute score of target susceptibility user and its corresponding weighted value
Power summation, obtains the value-at-risk of the target susceptibility user；Wherein, any in all sensitive users of target susceptibility user
One sensitive users.
Optionally, the user behaviors log information includes title, Webpage search information and the user network that user accesses website
Release information.
The embodiment of the invention also provides a kind of network behavior monitoring methods, comprising:
The user behaviors log information that will acquire is matched with preset sensitive keys words group, determines sensitive use
Family；Wherein, the user behaviors log information is generated by the network behavior of user；
According to the property parameters of all kinds of label informations, the value-at-risk of each sensitive users is determined；
The user behaviors log information for being more than the sensitive users of threshold value to value-at-risk is monitored；When appearance and the sensitive keys
When the matched user behaviors log information of words group, warning note is carried out.
Optionally, the user behaviors log information that will acquire is matched with preset sensitive keys words group, really
Making sensitive users includes:
It is filtered out from the user behaviors log information of acquisition specified using user behaviors log information corresponding with designated domain name；
Word segmentation processing is carried out to filtered user behaviors log information, obtains multiple participle groups；Wherein, each user has its right
The participle group answered；
Each participle group is matched with preset sensitive keys words group, determines sensitive users.
Optionally, described to match each participle group with preset sensitive keys words group, determine sensitive use
Family includes:
The number that the sensitive participle to match in statistics target participle group with sensitive keys words group occurs；Wherein, target
Participle group is any one participle group in all participle groups；
When the number that sensitive participle occurs is greater than or equal to preset threshold, then the corresponding use of the target participle group is determined
Family is sensitive users；
When the number that sensitive participle occurs is less than the preset threshold, then judge be in target histories user behaviors log information
It is no the sensitive participle occur；Wherein, the target histories user behaviors log information is the corresponding user's of the target participle group
The user behaviors log information that web-based history behavior generates；
When occurring the sensitive participle in target histories user behaviors log information, then the corresponding use of the target participle group is determined
Family is sensitive users.
Optionally, described to match each participle group with preset sensitive keys words group, determine sensitive use
Family includes:
The total number of the sensitive participle to match in statistics target participle group with sensitive keys words group；Wherein, target point
Phrase is any one participle group in all participle groups；
When the total number of sensitivity participle is greater than pre-set limit, then determine that the corresponding user of the target participle group is sensitivity
User.
Optionally, it is directed to the determination process of the property parameters of all kinds of label informations, which comprises
Obtain sample data；Wherein, the sample data includes being provided with the user behaviors log information and not of sensitivity label
The user behaviors log information of sensitivity label is set；
The detection model for being provided with all kinds of label informations is trained using the sample data, until the detection mould
The accuracy rate of type meets preset requirement, then by the corresponding attribute score of label informations all kinds of in trained detection model and weight
Value is used as property parameters.
Optionally, the property parameters according to all kinds of label informations determine that the value-at-risk of each sensitive users includes:
The user behaviors log information of each sensitive users is matched with all kinds of label informations, determines each of each sensitive users
Item attribute score；Wherein, every class label packet contains at least one attribute information, and each attribute information has its corresponding attribute
Score；And every class label information has its corresponding weighted value；
Every attribute score of target susceptibility user and its corresponding weighted value are weighted summation, obtain the target
The value-at-risk of sensitive users；Wherein, any one sensitive users in all sensitive users of target susceptibility user.
Optionally, the user behaviors log information includes title, Webpage search information and the user network that user accesses website
Release information.
The embodiment of the invention also provides a kind of network behavior monitoring devices, comprising:
Memory, for storing computer program；
Processor, for executing the computer program with realize the user behaviors log information that will acquire with it is preset quick
Sense keyword phrases are matched, and determine sensitive users；Wherein, the user behaviors log information is produced by the network behavior of user
It is raw；According to the property parameters of all kinds of label informations, the value-at-risk of each sensitive users is determined；It is more than the sensitivity of threshold value to value-at-risk
The user behaviors log information of user is monitored；When there is user behaviors log information matched with the sensitive keys words group, into
The step of row warning note.
The embodiment of the invention also provides a kind of computer readable storage medium, deposited on the computer readable storage medium
Computer program is contained,
The user behaviors log information and preset sensitivity that will acquire are realized when the computer program is executed by processor
Keyword phrases are matched, and determine sensitive users；Wherein, the user behaviors log information is generated by the network behavior of user；
According to the property parameters of all kinds of label informations, the value-at-risk of each sensitive users is determined；It is more than that the sensitive of threshold value is used to value-at-risk
The user behaviors log information at family is monitored；When there is user behaviors log information matched with the sensitive keys words group, carry out
The step of warning note.
Network behavior monitoring device includes matching unit, determination unit, tracing unit it can be seen from above-mentioned technical proposal
And Alarm Unit；The user behaviors log information and the progress of preset sensitive keys words group that matching unit is used to will acquire
Match, determines sensitive users；Wherein, user behaviors log information is generated by the network behavior of user；Determination unit, for according to all kinds of
The property parameters of label information determine the value-at-risk of each sensitive users；Label information can be according to the historical behavior of each user
The information attribute that log information includes divides obtain in advance.Value-at-risk is higher, illustrates that sensitive users make the probability of abnormal behaviour
It is bigger.In order to reduce the generation of sensitive users fortuitous event, the tracing unit of network behavior monitoring device can use to risk
Value is more than that the user behaviors log information of the sensitive users of threshold value is monitored, in order to find the abnormal behaviour of sensitive users in time.
When user behaviors log information matched with sensitive keys words group occur again in the sensitive users that value-at-risk is more than threshold value, then pass through
Alarm Unit carries out warning note, so that notification event processing people quickly intervenes solution, effectively avoids the generation of fortuitous event.
Detailed description of the invention
In order to illustrate the embodiments of the present invention more clearly, attached drawing needed in the embodiment will be done simply below
It introduces, it should be apparent that, drawings in the following description are only some embodiments of the invention, for ordinary skill people
For member, without creative efforts, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of structural schematic diagram of network behavior monitoring device provided in an embodiment of the present invention；
Fig. 2 is a kind of flow chart of network behavior monitoring method provided in an embodiment of the present invention；
Fig. 3 is a kind of hardware structural diagram of network behavior monitoring equipment provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, rather than whole embodiments.Based on this
Embodiment in invention, those of ordinary skill in the art are without making creative work, obtained every other
Embodiment belongs to the scope of the present invention.
In order to enable those skilled in the art to better understand the solution of the present invention, with reference to the accompanying drawings and detailed description
The present invention is described in further detail.
Next, a kind of network behavior monitoring device provided by the embodiment of the present invention is discussed in detail.Fig. 1 is that the present invention is real
A kind of structural schematic diagram of network behavior monitoring device of example offer is applied, which includes matching unit 11, determination unit 12, chases after
Track unit 13 and Alarm Unit 14；
Matching unit 11, user behaviors log information and the progress of preset sensitive keys words group for will acquire
Match, determines sensitive users；Wherein, user behaviors log information is generated by the network behavior of user.
The network behavior of user may include user surf the web information, by Web Publishing speech, the transmission row such as mail
For.Correspondingly, may include title, the webpage that user accesses website according to the user behaviors log information that the network behavior of user generates
Search for information and user network release information etc..Wherein, user network release information may include user post microblogging, publication
Comment, pass through chat content, mail of dealing of Internet chat tool etc..
In internet system field, it is provided with the internet behavior audit for being monitored to each user network behavior and sets
Standby, which can collect the user behaviors log information that the network behavior of each user generates.Implement in the present invention
In example, network behavior monitoring device be may be mounted in internet behavior audit device, can also be carried by other hardware devices.
In view of having usually contained some sensitive vocabulary in the user behaviors log information of sensitive users, for example, having suicide to incline
To sensitive group user behaviors log information in may include charcoal-burning suicide, purchase hypnotic or the sensitive vocabulary such as not would like to live.
Therefore, in embodiments of the present invention, in order to quickly find sensitive users present in the network user, can preset quick
Feel keyword phrases.In practical applications, it can therefrom be extracted by the user behaviors log information of a large amount of sensitive users of collection
The sensitive words often occurred constitute sensitive keys words group.
When occurring the sensitive words to match with sensitive keys words group in the user behaviors log information of user, then explanation should
User belongs to sensitive users.In order to further determine the degree of risk of sensitive users, 12 pairs of determination unit sensitive use can be passed through
It is further assessed at family.
Determination unit 12 determines the value-at-risk of each sensitive users for the property parameters according to all kinds of label informations.
Determination unit 12 is used to carry out quantitative evaluation to the network behavior of sensitive users.The value-at-risk of sensitive users is bigger,
Then illustrate that the sensitive users are easier to make injury oneself or injure other people the even social abnormal behaviours of harm.
In embodiments of the present invention, the information category that label information can include according to the historical behavior log information of each user
Property divide to obtain, for example, divide attribute may include role, profession, gender, access uniform resource locator (Uniform
Resource Locator, URL) classification, search content, access time etc..
Every attribute can be used as a kind of label information, wherein every class label packet contains at least one attribute information.
In view of the significance level of each attribute information is different, its corresponding attribute score can be set for each attribute information.It is different
Label information it is different to the disturbance degree of the value-at-risks of sensitive users assessment, therefore be directed to every class label information and it can be set
Corresponding weighted value.
The value of property parameters is the important parameter for influencing value-at-risk and calculating, in order to guarantee the reasonable of property parameters setting
Property, in practical applications, the property parameters of all kinds of label informations can be determined by way of sample training.
In embodiments of the present invention, the determination process of the property parameters of all kinds of label informations, network behavior monitoring are directed to
Device further includes acquiring unit, training unit and as unit；
Acquiring unit, for obtaining sample data；Wherein, sample data includes the user behaviors log letter for being provided with sensitivity label
The user behaviors log information of breath and not set sensitivity label.
In practical applications, user behaviors log can be believed in conjunction with the performance of the daily behavior of sensitive keys words group and user
Breath is marked.Such as the sensitive group with introgression, sensitive people is determined by sensitive keys word phrase matching
After group, we can for the behavior expression in sensitive group its recent real life recognized, further confirm that whether
Sensitivity label is carried out to the user behaviors log information of sensitive group.Specifically, can be come by carrying out psychological test to sensitive group
Assess the behavior expression of its real life.If there also have behavior to occur in real life to be abnormal, sensitivity label is set.
For example, the sensitive group with introgression recognized there are 50 in 50000 people, can be sent out to personnel
Play psychological research questionnaire, but 50000 parts of questionnaires collected, we only need to be for being identified as 50 of sensitive group
People carries out questionnaire analysis, and other staff can directly ignore, and can greatly reduce the time of analysis in this way.It is logical
The daily behavior performance in conjunction with this 50 personnel is crossed, it is real high risk sensitive group which can go out with comprehensive judgement, thus
Sensitivity label is arranged in the user behaviors log information of high risk sensitive group.
Training unit, for being trained using sample data to the detection model for being provided with all kinds of label informations, until
The accuracy rate of detection model meets preset requirement, then triggers as unit；It can will be in trained detection model as unit
The corresponding attribute score of all kinds of label informations and weighted value are as property parameters.It, can basis when subsequent value-at-risk calculates
The attribute score and weighted value that training obtains, calculate the value-at-risk of sensitive users.
The training process of detection model is the mistake constantly adjusted to the corresponding attribute score of all kinds of label informations and weighted value
Journey.When the accuracy rate of detection model meets preset requirement, then it is optimal to illustrate that the property parameters of the detection model have been adjusted to
Value.
Include role, profession, gender, access URL classification, search content and for access time by label information, passes through
Detection model is trained, can determine the corresponding attribute score of label information and weighted value.The attribute for including such as role
Information can have: undergraduate, postgraduate, doctor etc., and corresponding attribute score is followed successively by 5,6,7；The corresponding weight of role
Value is 0.2.The attribute information that profession includes can have: computer, mathematics, medicine etc., and corresponding attribute score is followed successively by 6,
5,7；The corresponding weighted value of profession is 0.3.The attribute information that gender includes has: male, female, corresponding attribute score are followed successively by 4,
6；The corresponding weighted value of gender is 0.3.The attribute information that access url classification includes can have: religion, news, and amusement is pornographic
Deng corresponding attribute score is followed successively by 4,2,3,6；Accessing the corresponding weighted value of url classification is 0.8.Searching for content includes
Attribute information can have: charcoal-burning suicide, buy hypnotic, not would like to live, corresponding attribute score is followed successively by 6,7,9；It searches
The corresponding weighted value of rope content is 0.9.The attribute information that access time includes can have: daytime, and at night, corresponding attribute obtains
Divide and is followed successively by 5,8；Access time corresponding weighted value is 0.4.
It should be noted that the example above illustrates it is only enumerating for simple label information and attribute information, in reality
In, the type of the label information of setting can have more kinds of, and attribute information described under every class label information can be with
There are more.
It is directed to the calculating of value-at-risk, determination unit 12 may include coupling subelement and summation subelement.
Coupling subelement is determined for matching the user behaviors log information of each sensitive users with all kinds of label informations
Every attribute score of each sensitive users out.
Sensitive users may have multiple, and the calculation of the value-at-risk of each sensitive users is identical, with all sensitive users
In any one sensitive users, that is, target susceptibility user for be introduced.
Summation subelement, for every attribute score of target susceptibility user and its corresponding weighted value to be weighted and ask
With obtain the value-at-risk of target susceptibility user.
The value-at-risk R calculation formula of sensitive users is as follows:
R=(k1*x1+k2*x2+k3*x3+...kn*xn)/n；
Wherein, kn indicates the weighted value of the n-th class label information；Xn indicates that sensitive users attribute is believed under the n-th class label information
The attribute score of breath, n indicate the total number of attribute information.
For example, a sensitive users role is postgraduate, profession is medicine, and gender female, access url classification is pornographic, is searched
Rope content includes " not would like to live " this attribute information, and access time a bit, believes for morning in conjunction with labels all kinds of in the example above
Corresponding attribute score and weighted value are ceased, can determine that the value-at-risk of the sensitive users is R=(0.2*6+0.3*7+0.3*
6+0.8*6+0.9*9+0.4*8)/6=3.5.
If sensitive users search content contains two attribute informations simultaneously, such as " charcoal-burning suicide " and " not would like to live
", then the calculation of its value-at-risk are as follows:
R=(0.2*6+0.3*7+0.3*6+0.8*6+0.9*6+0.9*9+0.4*8)/7=3.8.
When the value-at-risk of sensitive users is more than threshold value, then illustrates that the sensitive users very likely make abnormal behaviour, be
Convenient for finding that the Novel presentation of sensitive users in embodiments of the present invention can be by network behavior monitoring device in time
The user behaviors log information of sensitive users that is more than threshold value to value-at-risk of tracing unit 13 monitored in real time.Once occur with
When the matched user behaviors log information of sensitive keys words group, then triggers Alarm Unit 14 and carry out warning note.
Illustrate that the calculation of risk value, threshold value can be set to 3 in conjunction with the example above.It is more than 3 when there is value-at-risk
Sensitive users when, then the user behaviors log information of the sensitive users is monitored in real time.
In embodiments of the present invention, the mode of warning note can there are many, such as send out to specified event handling personnel
Send short message, wechat or mail etc..
For the sensitive group of high risk, alerted once triggering sensitive keys words group to corresponding event handling people
Member, in order to which event handling personnel take corresponding Strategies for action to avoid adverse events.For example, the height with introgression
Risk sensitivity personnel, if the period in morning, by chat mode, sending out help information, " hope has been can't see in life
, I should be what if I wants to leave this world " etc. this type of information, can trigger immediately at this time alarm to event handling
Personnel, event handling personnel, which take an immediate action, finds the sensitive personnel of the high risk, accomplishes to find ahead of time, effectively avoid as far as possible
The generation of accident.
Network behavior monitoring device includes matching unit, determination unit, tracing unit it can be seen from above-mentioned technical proposal
And Alarm Unit；The user behaviors log information and the progress of preset sensitive keys words group that matching unit is used to will acquire
Match, determines sensitive users；Wherein, user behaviors log information is generated by the network behavior of user；Determination unit, for according to all kinds of
The property parameters of label information determine the value-at-risk of each sensitive users；Label information can be according to the historical behavior of each user
The information attribute that log information includes divides obtain in advance.Value-at-risk is higher, illustrates that sensitive users make the probability of abnormal behaviour
It is bigger.In order to reduce the generation of sensitive users fortuitous event, the tracing unit of network behavior monitoring device can use to risk
Value is more than that the user behaviors log information of the sensitive users of threshold value is monitored, in order to find the abnormal behaviour of sensitive users in time.
When user behaviors log information matched with sensitive keys words group occur again in the sensitive users that value-at-risk is more than threshold value, then pass through
Alarm Unit carries out warning note, so that notification event processing people quickly intervenes solution, effectively avoids the generation of fortuitous event.
Matching unit 11 determines sensitive users for matching user behaviors log information with sensitive keys words group.
In practical applications, can using participle technique to user behaviors log information carry out word segmentation processing, by obtained multiple participles with it is quick
Sense keyword phrases are matched.
The matched accuracy of sensitive keys words group can be promoted using participle technique, reduce similar " material flow industry " meeting
The probability that the erroneous judgement situation such as be judged to " miscarrying " occurs.
In embodiments of the present invention, it is contemplated that may in the social news of some regular websites or news portal report
Also it can be related to some sensitive vocabulary, but these sensitive vocabulary describe media event, can not reflect the behavior table of user
It is existing, therefore, in order to further enhance the matched accuracy of sensitive keys words group, filtering function can be set.
Specifically, matching unit 11 may include filtering subelement, participle subelement and determine subelement.
Subelement is filtered, it is specified using row corresponding with designated domain name for being filtered out from the user behaviors log information of acquisition
For log information.
Specified application may include regular news portal, such as Tencent's news, Baidu's news.Designated domain name may include
Regular web page address information.Other than filtering out specified application log information corresponding with designated domain name, in practical application
In other types of secure data can also be filtered, it is not limited here according to demand.
Before carrying out keyword phrases matching, specified application and specified domain can will be accessed in User action log information
Log information corresponding to name is deleted, to avoid containing sensitive vocabulary in the common access behaviors such as news access and influencing
Matched accuracy rate.
Subelement is segmented, for carrying out word segmentation processing to filtered user behaviors log information, obtains multiple participle groups；Its
In, each user has its corresponding participle group.
It determines subelement, for matching each participle group with preset sensitive keys words group, determines quick
Feel user.
It, both can be according to the sensitivity for including in user's current behavior log information when determining sensitive users in practical applications
The total number of vocabulary determines whether user is sensitive users.It can also be by user's current behavior log information and target histories
The distribution situation of sensitive vocabulary in user behaviors log information, determines whether user is sensitive users.
Each user has its corresponding participle group, each user whether be sensitive users decision procedure it is similar, In
During the present invention is implemented, it is unfolded to introduce by taking any one participle group, that is, target participle group in all participle groups as an example.
For in the first manner, determine that subelement matches for counting in target participle group with sensitive keys words group
Sensitive participle total number；When the total number of sensitivity participle is greater than pre-set limit, then the corresponding use of target participle group is determined
Family is sensitive users.
For in the second, determine that subelement matches for counting in target participle group with sensitive keys words group
The number that occurs of sensitive participle.
It is possible that multiple and different sensitive participles, each sensitive participle have its corresponding occurrence out in target participle group
Number.When the number that the same sensitive participle occurs is greater than or equal to preset threshold, then the corresponding user of target participle group is determined
For sensitive users.
When the number that sensitive participle each in target participle group occurs is respectively less than preset threshold, in order to determine user whether be
Sensitive users can further judge whether occur sensitivity identical with the target participle group in target histories user behaviors log information
Participle；Wherein, target histories user behaviors log information is the behavior of the web-based history behavior generation of the corresponding user of target participle group
Log information.
When occurring sensitive participle identical with the target participle group in target histories user behaviors log information, then illustrate user
Last longer pays close attention to such sensitive participle, can be determined that the corresponding user of target participle group is quick at this time
Feel user.
It, can be more quasi- by combining the target histories user behaviors log information of user to assess the network behavior of user
True identifies the network behavior of user, effectively reduces the omission of sensitive group detection, improves sensitive group mirror
Other accuracy.
It is illustrated in figure 2 a kind of flow chart of network behavior monitoring method provided in an embodiment of the present invention, this method comprises:
S201: the user behaviors log information that will acquire is matched with preset sensitive keys words group, is determined quick
Feel user.
Wherein, user behaviors log information is generated by the network behavior of user.
S202: according to the property parameters of all kinds of label informations, the value-at-risk of each sensitive users is determined.
Wherein, label information divides to obtain according to the information attribute that the historical behavior log information of each user includes.
Property parameters may include attribute score and weighted value.Every class label information includes having multinomial attribute information, each
Attribute information has its corresponding attribute score, and every class label information has its corresponding weighted value.
S203: the user behaviors log information for being more than the sensitive users of threshold value to value-at-risk monitors in real time；When occur with it is quick
When feeling the matched user behaviors log information of keyword phrases, warning note is carried out.
Optionally, the user behaviors log information that will acquire is matched with preset sensitive keys words group, is determined
Sensitive users include:
It is filtered out from the user behaviors log information of acquisition specified using user behaviors log information corresponding with designated domain name；
Word segmentation processing is carried out to filtered user behaviors log information, obtains multiple participle groups；Wherein, each user has its right
The participle group answered；
Each participle group is matched with preset sensitive keys words group, determines sensitive users.
Optionally, each participle group is matched with preset sensitive keys words group, determines sensitive users packet
It includes:
The number that the sensitive participle to match in statistics target participle group with sensitive keys words group occurs；Wherein, target
Participle group is any one participle group in all participle groups；
When the number that sensitive participle occurs is greater than or equal to preset threshold, then determine that the corresponding user of target participle group is
Sensitive users；
When the number that sensitive participle occurs is less than preset threshold, then judge whether go out in target histories user behaviors log information
Existing sensitive participle；Wherein, target histories user behaviors log information is that the web-based history behavior of the corresponding user of target participle group generates
User behaviors log information；
When occurring sensitive participle in target histories user behaviors log information, then determine that the corresponding user of target participle group is sensitivity
User.
Optionally, each participle group is matched with preset sensitive keys words group, determines sensitive users packet
It includes:
The total number of the sensitive participle to match in statistics target participle group with sensitive keys words group；Wherein, target point
Phrase is any one participle group in all participle groups；
When the total number of sensitivity participle is greater than pre-set limit, then determine that the corresponding user of target participle group uses for sensitivity
Family.
Optionally, it is directed to the determination process of the property parameters of all kinds of label informations, method includes:
Obtain sample data；Wherein, sample data includes being provided with the user behaviors log information of sensitivity label and not set
The user behaviors log information of sensitivity label；
The detection model for being provided with all kinds of label informations is trained using sample data, until detection model is accurate
Rate meets preset requirement, then using the corresponding attribute score of label informations all kinds of in trained detection model and weighted value as category
Property parameter.
Optionally, according to the property parameters of all kinds of label informations, determine that the value-at-risk of each sensitive users includes:
The user behaviors log information of each sensitive users is matched with all kinds of label informations, determines each of each sensitive users
Item attribute score；Wherein, every class label packet contains at least one attribute information, and each attribute information has its corresponding attribute
Score；And every class label information has its corresponding weighted value；
Every attribute score of target susceptibility user and its corresponding weighted value are weighted summation, obtain target susceptibility
The value-at-risk of user；Wherein, any one sensitive users in all sensitive users of target susceptibility user.
Optionally, user behaviors log information includes title, Webpage search information and the user network publication that user accesses website
Information.
The explanation of feature may refer to the related description of embodiment corresponding to Fig. 1 in embodiment corresponding to Fig. 2, here no longer
It repeats one by one.
The user behaviors log information and preset sensitive keys words group that will acquire it can be seen from above-mentioned technical proposal
It is matched, determines sensitive users；Wherein, user behaviors log information is generated by the network behavior of user；Believed according to all kinds of labels
The property parameters of breath determine the value-at-risk of each sensitive users；Wherein, label information is believed according to the historical behavior log of each user
The information attribute that breath includes can divide to obtain.Value-at-risk is higher, illustrate sensitive users make abnormal behaviour probability it is bigger.For
The generation of sensitive users fortuitous event is reduced, can be more than that the user behaviors log information of the sensitive users of threshold value is carried out to value-at-risk
Monitoring, in order to find the abnormal behaviour of sensitive users in time.When value-at-risk be more than threshold value sensitive users occur again with it is quick
When feeling the matched user behaviors log information of keyword phrases, then warning note is carried out, so that notification event processing people quickly intervenes solution
Certainly, the generation of fortuitous event is effectively avoided.
It is illustrated in figure 3 a kind of hardware structural diagram of network behavior monitoring device 30 provided in an embodiment of the present invention,
Include:
Memory 31, for storing computer program；
Processor 32, for executing computer program to realize the user behaviors log information that will acquire and preset sensitivity
Keyword phrases are matched, and determine sensitive users；Wherein, user behaviors log information is generated by the network behavior of user；According to
The property parameters of all kinds of label informations determine the value-at-risk of each sensitive users；Wherein, label information is according to the history of each user
The information attribute that user behaviors log information includes divides to obtain；To value-at-risk be more than threshold value sensitive users user behaviors log information into
Row real time monitoring；When occurring with the matched user behaviors log information of sensitive keys words group, the step of carrying out warning note.
The embodiment of the invention also provides a kind of computer readable storage medium, it is stored on computer readable storage medium
Computer program,
The user behaviors log information that will acquire and preset sensitive keys are realized when computer program is executed by processor
Words group is matched, and determines sensitive users；Wherein, user behaviors log information is generated by the network behavior of user；According to all kinds of
The property parameters of label information determine the value-at-risk of each sensitive users；Wherein, label information is according to the historical behavior of each user
The information attribute that log information includes divides to obtain；The user behaviors log information that value-at-risk is more than the sensitive users of threshold value is carried out real
When monitor；When occurring with the matched user behaviors log information of sensitive keys words group, the step of carrying out warning note.
It is provided for the embodiments of the invention a kind of network behavior monitoring device, method, equipment and computer-readable above
Storage medium is described in detail.Each embodiment is described in a progressive manner in specification, and each embodiment emphasis is said
Bright is the difference from other embodiments, and the same or similar parts in each embodiment may refer to each other.For reality
For applying method disclosed in example, since it is corresponding with device disclosed in embodiment, so being described relatively simple, related place
Illustrate referring to device part.It should be pointed out that for those skilled in the art, not departing from the present invention
, can be with several improvements and modifications are made to the present invention under the premise of principle, these improvement and modification also fall into right of the present invention
It is required that protection scope in.
Professional further appreciates that, unit described in conjunction with the examples disclosed in the embodiments of the present disclosure
And algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These
Function is implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Profession
Technical staff can use different methods to achieve the described function each specific application, but this realization is not answered
Think beyond the scope of this invention.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
Claims (10)
1. a kind of network behavior monitoring device, which is characterized in that single including matching unit, determination unit, tracing unit and alarm
Member；
The matching unit, the user behaviors log information for will acquire are matched with preset sensitive keys words group,
Determine sensitive users；Wherein, the user behaviors log information is generated by the network behavior of user；
The determination unit determines the value-at-risk of each sensitive users for the property parameters according to all kinds of label informations；Its
In, the label information divides to obtain according to the information attribute that the historical behavior log information of each user includes；
The tracing unit is monitored for the user behaviors log information to value-at-risk more than the sensitive users of threshold value；
The Alarm Unit, for alarming when there is user behaviors log information matched with the sensitive keys words group
Prompt.
2. the apparatus according to claim 1, which is characterized in that the matching unit includes filtering subelement, participle list
Member and determining subelement；
The filtering subelement, it is specified using row corresponding with designated domain name for being filtered out from the user behaviors log information of acquisition
For log information；
The participle subelement obtains multiple participle groups for carrying out word segmentation processing to filtered user behaviors log information；Its
In, each user has its corresponding participle group；
The determining subelement is determined quick for matching each participle group with preset sensitive keys words group
Feel user.
3. the apparatus according to claim 1, which is characterized in that the determining subelement for count in target participle group with
The number that the sensitive participle that sensitive keys words group matches occurs；Wherein, target participle group is any in all participle groups
One participle group；
When the number that sensitive participle occurs is greater than or equal to preset threshold, then determine that the corresponding user of the target participle group is
Sensitive users；
When the number that sensitive participle occurs is less than the preset threshold, then judge whether go out in target histories user behaviors log information
The existing sensitive participle；Wherein, the target histories user behaviors log information is the history of the corresponding user of the target participle group
The user behaviors log information that network behavior generates；
When occurring the sensitive participle in target histories user behaviors log information, then determine that the corresponding user of the target participle group is
Sensitive users.
4. the apparatus according to claim 1, which is characterized in that the determining subelement for count in target participle group with
The total number for the sensitive participle that sensitive keys words group matches；Wherein, target participle group is any one in all participle groups
A participle group；
When the total number of sensitivity participle is greater than pre-set limit, then determine that the corresponding user of the target participle group uses for sensitivity
Family.
5. the apparatus according to claim 1, which is characterized in that be directed to the property parameters of all kinds of label informations really
Determine process, described device further includes acquiring unit, training unit and as unit；
The acquiring unit, for obtaining sample data；Wherein, the sample data includes being provided with the behavior day of sensitivity label
The user behaviors log information of will information and not set sensitivity label；
The training unit, for being trained using the sample data to the detection model for being provided with all kinds of label informations,
Until the accuracy rate of the detection model meets preset requirement, then trigger described as unit；
It is described to be used as unit, for the corresponding attribute score of label informations all kinds of in trained detection model and weighted value to be made
For property parameters.
6. device according to claim 5, which is characterized in that the determination unit includes that coupling subelement and summation are single
Member；
The coupling subelement is determined for matching the user behaviors log information of each sensitive users with all kinds of label informations
Every attribute score of each sensitive users out；Wherein, every class label packet contains at least one attribute information, each attribute letter
Breath has its corresponding attribute score；And every class label information has its corresponding weighted value；
The summation subelement, for every attribute score of target susceptibility user and its corresponding weighted value to be weighted and ask
With obtain the value-at-risk of the target susceptibility user；Wherein, any one in all sensitive users of target susceptibility user
Sensitive users.
7. device described in -6 any one according to claim 1, which is characterized in that the user behaviors log information includes user's visit
Ask title, Webpage search information and the user network release information of website.
8. a kind of network behavior monitoring method characterized by comprising
The user behaviors log information that will acquire is matched with preset sensitive keys words group, determines sensitive users；Its
In, the user behaviors log information is generated by the network behavior of user；
According to the property parameters of all kinds of label informations, the value-at-risk of each sensitive users is determined；Wherein, the label information according to
The information attribute that the historical behavior log information of each user includes divides to obtain；
The user behaviors log information for being more than the sensitive users of threshold value to value-at-risk is monitored；When appearance and the sensitive keys words
When the matched user behaviors log information of group, warning note is carried out.
9. a kind of network behavior monitoring device characterized by comprising
Memory, for storing computer program；
Processor, for executing the computer program to realize the user behaviors log information that will acquire and preset sensitive pass
Key words group is matched, and determines sensitive users；Wherein, the user behaviors log information is generated by the network behavior of user；Root
According to the property parameters of all kinds of label informations, the value-at-risk of each sensitive users is determined；Wherein, the label information is according to each user
The historical behavior log information information attribute that includes divide to obtain；It is more than the user behaviors log of the sensitive users of threshold value to value-at-risk
Information is monitored；When there is user behaviors log information matched with the sensitive keys words group, the step of warning note is carried out
Suddenly.
10. a kind of computer readable storage medium, which is characterized in that be stored with computer on the computer readable storage medium
Program,
The user behaviors log information that will acquire and preset sensitive keys are realized when the computer program is executed by processor
Words group is matched, and determines sensitive users；Wherein, the user behaviors log information is generated by the network behavior of user；According to
The property parameters of all kinds of label informations determine the value-at-risk of each sensitive users；Wherein, the label information is according to each user's
The information attribute that historical behavior log information includes divides to obtain；It is more than the user behaviors log letter of the sensitive users of threshold value to value-at-risk
Breath is monitored；When occurring with the matched user behaviors log information of the sensitive keys words group, the step of carrying out warning note.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910810292.8A CN110516156B (en) | 2019-08-29 | 2019-08-29 | Network behavior monitoring device, method, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910810292.8A CN110516156B (en) | 2019-08-29 | 2019-08-29 | Network behavior monitoring device, method, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110516156A true CN110516156A (en) | 2019-11-29 |
| CN110516156B CN110516156B (en) | 2023-03-17 |
Family
ID=68629160
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910810292.8A Active CN110516156B (en) | 2019-08-29 | 2019-08-29 | Network behavior monitoring device, method, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110516156B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111045910A (en) * | 2019-12-13 | 2020-04-21 | 中国电信集团工会上海市委员会 | User attribute identification method based on big data |
| CN111079029A (en) * | 2019-12-20 | 2020-04-28 | 珠海格力电器股份有限公司 | Sensitive account detection method, storage medium and computer equipment |
| CN111629235A (en) * | 2020-01-20 | 2020-09-04 | 全息空间(深圳)智能科技有限公司 | Live broadcast platform user data analysis method and device and storage medium |
| CN111666317A (en) * | 2020-07-06 | 2020-09-15 | 腾讯科技（深圳）有限公司 | Cheating information mining method and cheating information identification method and device |
| CN111930977A (en) * | 2020-08-04 | 2020-11-13 | 腾讯音乐娱乐科技（深圳）有限公司 | Method, device and equipment for identifying sensitive information and storage medium |
| CN112965979A (en) * | 2021-03-10 | 2021-06-15 | 中国民航信息网络股份有限公司 | User behavior analysis method and device and electronic equipment |
| CN113254761A (en) * | 2021-01-04 | 2021-08-13 | 北京汇达城数科技发展有限公司 | Intelligent early warning system and method for specific behavior information |
| CN113891120A (en) * | 2021-09-29 | 2022-01-04 | 广东省高峰科技有限公司 | IPTV service terminal access method and system thereof |
| CN115827414A (en) * | 2023-02-15 | 2023-03-21 | 天津戎行集团有限公司 | Network user behavior monitoring and analyzing method based on open source data |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012221383A (en) * | 2011-04-12 | 2012-11-12 | Nippon Telegr & Teleph Corp <Ntt> | Mental health management support system and mental health management support method |
| CN106411578A (en) * | 2016-09-12 | 2017-02-15 | 国网山东省电力公司电力科学研究院 | Website monitoring system and method applicable to power industry |
| US20170103200A1 (en) * | 2015-10-07 | 2017-04-13 | Soliton Systems, K.K. | Log Information Generation Apparatus And Recording Medium, And Log Information Extraction Apparatus And Recording Medium |
| US20170251007A1 (en) * | 2016-02-29 | 2017-08-31 | Soliton Systems K.K. | Automated computer behavioral analysis system and methods |
| CN107391680A (en) * | 2017-07-24 | 2017-11-24 | 北京京东尚科信息技术有限公司 | Content recommendation method, device and equipment |
| CN107943954A (en) * | 2017-11-24 | 2018-04-20 | 杭州安恒信息技术有限公司 | Detection method, device and the electronic equipment of webpage sensitive information |
| CN107995289A (en) * | 2017-12-01 | 2018-05-04 | 中国联合网络通信集团有限公司 | The behaviortrace method and platform of a kind of network user |
| CN108153760A (en) * | 2016-12-05 | 2018-06-12 | 中国移动通信有限公司研究院 | Network gambling monitoring method and device |
| CN109284441A (en) * | 2018-09-18 | 2019-01-29 | 中国人民解放军战略支援部队信息工程大学 | Dynamic self-adapting network sensitive information detection method and device |
| CN109558480A (en) * | 2018-11-30 | 2019-04-02 | 重庆市千将软件有限公司 | For the counter method of crime of laundering behavior |
| CN110096408A (en) * | 2019-03-11 | 2019-08-06 | 中国平安人寿保险股份有限公司 | Alarm-monitor method, apparatus, electronic equipment and computer readable storage medium |
-
2019
- 2019-08-29 CN CN201910810292.8A patent/CN110516156B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012221383A (en) * | 2011-04-12 | 2012-11-12 | Nippon Telegr & Teleph Corp <Ntt> | Mental health management support system and mental health management support method |
| US20170103200A1 (en) * | 2015-10-07 | 2017-04-13 | Soliton Systems, K.K. | Log Information Generation Apparatus And Recording Medium, And Log Information Extraction Apparatus And Recording Medium |
| US20170251007A1 (en) * | 2016-02-29 | 2017-08-31 | Soliton Systems K.K. | Automated computer behavioral analysis system and methods |
| CN106411578A (en) * | 2016-09-12 | 2017-02-15 | 国网山东省电力公司电力科学研究院 | Website monitoring system and method applicable to power industry |
| CN108153760A (en) * | 2016-12-05 | 2018-06-12 | 中国移动通信有限公司研究院 | Network gambling monitoring method and device |
| CN107391680A (en) * | 2017-07-24 | 2017-11-24 | 北京京东尚科信息技术有限公司 | Content recommendation method, device and equipment |
| CN107943954A (en) * | 2017-11-24 | 2018-04-20 | 杭州安恒信息技术有限公司 | Detection method, device and the electronic equipment of webpage sensitive information |
| CN107995289A (en) * | 2017-12-01 | 2018-05-04 | 中国联合网络通信集团有限公司 | The behaviortrace method and platform of a kind of network user |
| CN109284441A (en) * | 2018-09-18 | 2019-01-29 | 中国人民解放军战略支援部队信息工程大学 | Dynamic self-adapting network sensitive information detection method and device |
| CN109558480A (en) * | 2018-11-30 | 2019-04-02 | 重庆市千将软件有限公司 | For the counter method of crime of laundering behavior |
| CN110096408A (en) * | 2019-03-11 | 2019-08-06 | 中国平安人寿保险股份有限公司 | Alarm-monitor method, apparatus, electronic equipment and computer readable storage medium |
Non-Patent Citations (5)
| Title |
|---|
| CARLO BERNASCHINA 等: "Integrating Modeling Languages and Web Logs for Enhanced User Behavior Analytics", 《ACM》 * |
| 吴行惠等: "质监大数据企业画像的研究及应用", 《数字通信世界》 * |
| 李栋科: "一种面向海量网络审计日志的敏感用户挖掘分析架构", 《网络空间安全》 * |
| 杨明: "《移动互联时代出版传媒业发展的机遇与挑战》", 28 February 2018 * |
| 黄振江: "网络数据实现对物联网终端网络行为监控的措施研究", 《计算机产品与流通》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111045910B (en) * | 2019-12-13 | 2023-12-01 | 中国电信集团工会上海市委员会 | User attribute identification method based on big data |
| CN111045910A (en) * | 2019-12-13 | 2020-04-21 | 中国电信集团工会上海市委员会 | User attribute identification method based on big data |
| CN111079029B (en) * | 2019-12-20 | 2023-11-21 | 珠海格力电器股份有限公司 | Sensitive account detection method, storage medium and computer equipment |
| CN111079029A (en) * | 2019-12-20 | 2020-04-28 | 珠海格力电器股份有限公司 | Sensitive account detection method, storage medium and computer equipment |
| CN111629235A (en) * | 2020-01-20 | 2020-09-04 | 全息空间(深圳)智能科技有限公司 | Live broadcast platform user data analysis method and device and storage medium |
| CN111666317A (en) * | 2020-07-06 | 2020-09-15 | 腾讯科技（深圳）有限公司 | Cheating information mining method and cheating information identification method and device |
| CN111930977A (en) * | 2020-08-04 | 2020-11-13 | 腾讯音乐娱乐科技（深圳）有限公司 | Method, device and equipment for identifying sensitive information and storage medium |
| CN113254761A (en) * | 2021-01-04 | 2021-08-13 | 北京汇达城数科技发展有限公司 | Intelligent early warning system and method for specific behavior information |
| CN112965979A (en) * | 2021-03-10 | 2021-06-15 | 中国民航信息网络股份有限公司 | User behavior analysis method and device and electronic equipment |
| CN112965979B (en) * | 2021-03-10 | 2024-05-03 | 中国民航信息网络股份有限公司 | User behavior analysis method and device and electronic equipment |
| CN113891120A (en) * | 2021-09-29 | 2022-01-04 | 广东省高峰科技有限公司 | IPTV service terminal access method and system thereof |
| CN115827414A (en) * | 2023-02-15 | 2023-03-21 | 天津戎行集团有限公司 | Network user behavior monitoring and analyzing method based on open source data |
| CN115827414B (en) * | 2023-02-15 | 2023-05-02 | 天津戎行集团有限公司 | Network user behavior monitoring and analyzing method based on open source data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110516156B (en) | 2023-03-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110516156A (en) | A kind of network behavior monitoring device, method, equipment and storage medium | |
| Biyani et al. | " 8 amazing secrets for getting more clicks": detecting clickbaits in news streams using article informality | |
| Liu et al. | Forecasting the presence and intensity of hostility on Instagram using linguistic and social features | |
| US20210182301A1 (en) | Apparatus, systems and methods for scoring and distributing the reliablity of online information | |
| WO2019227710A1 (en) | Network public opinion analysis method and apparatus, and computer-readable storage medium | |
| Stafford et al. | An evaluation of the effect of spam on twitter trending topics | |
| US8768861B2 (en) | Research mission identification | |
| JP6301966B2 (en) | DATA ANALYSIS SYSTEM, DATA ANALYSIS METHOD, DATA ANALYSIS PROGRAM, AND RECORDING MEDIUM OF THE PROGRAM | |
| Sharma et al. | Detecting hate speech and insults on social commentary using nlp and machine learning | |
| McCreadie et al. | Incident streams 2019: Actionable insights and how to find them | |
| WO2020101477A1 (en) | System and method for dynamic entity sentiment analysis | |
| CN107544988A (en) | A kind of method and apparatus for obtaining public sentiment data | |
| CN108109445A (en) | Teaching class feelings monitoring method | |
| CN112328857B (en) | Product knowledge aggregation method and device, computer equipment and storage medium | |
| CN111159763A (en) | System and method for analyzing portrait of law-related personnel group | |
| CN115274122A (en) | Health medical data management method, system, electronic device and storage medium | |
| Liu et al. | Epic30m: An epidemics corpus of over 30 million relevant tweets | |
| Noor et al. | Comparing toxicity across social media platforms for COVID-19 discourse | |
| Granskogen | Automatic detection of fake news in social media using contextual information | |
| Zhong et al. | Identification of opinion spammers using reviewer reputation and clustering analysis | |
| Oostdijk et al. | Shallow parsing for recognizing threats in Dutch tweets | |
| CN103294686A (en) | Method and system for identifying webpage spam user and spam webpage | |
| CN116450924A (en) | Analysis module and method of intelligent environment-friendly platform ecological environment public opinion monitoring system | |
| CN113468206A (en) | Data maintenance method, device, server, medium and product | |
| Marche et al. | Implementation of a multi-approach fake news detector and of a trust management model for news sources |
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 |