CN109766367A - Focus incident determines method and device, computer equipment and storage medium - Google Patents
Focus incident determines method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109766367A CN109766367A CN201710629638.5A CN201710629638A CN109766367A CN 109766367 A CN109766367 A CN 109766367A CN 201710629638 A CN201710629638 A CN 201710629638A CN 109766367 A CN109766367 A CN 109766367A
- Authority
- CN
- China
- Prior art keywords
- event
- sequence
- events
- value
- focus incident
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Debugging And Monitoring (AREA)
Abstract
A kind of focus incident determines method, apparatus, equipment and storage medium, and the method for one embodiment includes: each sequence of events obtained in history event data library, any one sequence of events includes the frequency of each event and each event；To any one sequence of events, according to the frequency of each event in the sequence of events, the event value parameter value of each event in the sequence of events is determined；Also determine the recent validity parameter value of event of each event in the sequence of events in the sequence of events；To any one event, according to event value parameter value of the event in each sequence of events, it determines the event value parameter total value of the event, and according to the recent validity parameter value of event of the event in each sequence of events, determines the recent validity total value of the event of the event；Focus incident is determined according to the recent validity total value of the event value parameter total value and event of each event.Embodiment improves the accuracys and performance of focus incident prediction.
Description
Technical field
The present invention relates to digital digging technology fields, determine method, a kind of hot spot thing more particularly to a kind of focus incident
Part determining device, a kind of computer equipment and a kind of computer storage medium.
Background technique
Network hotspot event generally has sudden and interim, discovery (the Network Hot of network hotspot event
Events Mining, NHEM) it is referred to as topic detection and tracking (Topic detection and tracking, abbreviation
TDT), at present to the discovery and excavation of network hotspot event, data mining generally is carried out using various clustering algorithms.?
In current Data Mining Mechanism, realize simply, but huge based on different types of clustering technique calculation amount, cluster result is good
Bad related with the order of addition of new document, sensibility is big, periodically will lead to focus incident excavation applying to focus incident really
As a result inaccuracy.
Summary of the invention
Based on this, it is necessary to provide a kind of focus incident and determine method, a kind of focus incident determining device, a kind of computer
Equipment and a kind of computer storage medium excavate the accuracy for determining focus incident to improve.
Accordingly, following technical scheme is used in one embodiment:
A kind of focus incident determines method, comprising steps of
Each sequence of events in history event data library is obtained, any one sequence of events includes having time context
Each event and each event frequency；
The sequence of events is determined according to the frequency of each event in the sequence of events to any one sequence of events
In event value parameter value of each event in the sequence of events；Also according to the sequence of events in the history event data library
Sequence order, determine the recent validity parameter value of the sequence of the sequence of events, and have in the recent period according to the sequence of the sequence of events
Effect property parameter value, determines the recent validity parameter value of event of each event in the sequence of events in the sequence of events；
The event is determined according to event value parameter value of the event in each sequence of events to any one event
Event value parameter total value, and according to the recent validity parameter value of event of the event in each sequence of events, determine the event
The recent validity total value of event；
According to the recent validity total value of the event value parameter total value and event of each event, hot spot thing is determined from each event
Part.
A kind of focus incident determining device, comprising:
Module is obtained, for obtaining each sequence of events in history event data library, any one sequence of events includes tool
The frequency of the contextual each event of having time and each event；
It is worth determining module, for determining the event valence of each event in the sequence of events to any one sequence of events
Value parameter value；It is also used to any one event, according to event value parameter value of the event in each sequence of events, determining should
The event value parameter total value of event；
Recent validity determining module, for determining that the sequence of the sequence of events has in the recent period to any one sequence of events
Effect property parameter value, and determine event recent validity parameter value of each event in the sequence of events in the sequence of events；Also
For determining the event according to the recent validity parameter value of event of the event in each sequence of events to any one event
The recent validity total value of event；
Focus incident determining module, for total according to the recent validity of event value parameter total value and event of each event
Value determines focus incident from each event.
A kind of computer equipment, including memory, processor and be stored on the memory and can be in the processing
The computer program run on device, wherein the processor realizes method as described above when executing the computer program
Step.
A kind of computer storage medium, is stored thereon with computer program, wherein the realization when program is executed by processor
The step of method as described above.
Based on the scheme of embodiment as described above, based on contextual with the time in history event data library
The sequence of events of the frequency of each event and each event, by the event value parameter for determining each event in sequence of events
Value, and determine therefrom that the event value parameter total value of each event, and the recent validity ginseng of the sequence by determining each sequence of events
Numerical value determines therefrom that the recent validity total value of the event of each event, thus event value parameter total value and thing based on each event
The recent validity total value of part determines focus incident, thus by being carried out based on the sequence of events of temporal aspect to focus incident
It excavates and determines, breach the limitation that not can be carried out the focus incident for containing temporal aspect prediction, improve focus incident prediction
Accuracy and performance.
Detailed description of the invention
Fig. 1 is the schematic diagram of the working environment of a this embodiment scheme；
Fig. 2 is the schematic diagram of the composed structure of the terminal/server of one embodiment；
Fig. 3 is the flow diagram that focus incident in one embodiment determines method；
Fig. 4 is showing for the sequence tree that first sequence of events is inserted into an application example, reverse indexing table and look-up table
It is intended to；
Fig. 5 is showing for the sequence tree that second sequence of events is inserted into an application example, reverse indexing table and look-up table
It is intended to；
Fig. 6 is showing for the sequence tree that third sequence of events is inserted into an application example, reverse indexing table and look-up table
It is intended to；
Fig. 7 is showing for the sequence tree that the 4th sequence of events is inserted into an application example, reverse indexing table and look-up table
It is intended to；
Fig. 8 is showing for the sequence tree that the 5th sequence of events is inserted into an application example, reverse indexing table and look-up table
It is intended to；
Fig. 9 is the schematic diagram that the sequence tree after Frequent Subsequence is compressed in an application example；
Figure 10 is the schematic diagram for merging the sequence tree after a subsequence branch in an application example；
Figure 11 is the schematic diagram for merging the sequence tree after second subsequence branch in an application example；
Figure 12 is the schematic diagram for merging the sequence tree after third subsequence branch in an application example；
Figure 13 is to compress Frequent Subsequence in an application example and merge the schematic diagram of the sequence tree after subsequence branch；
Figure 14 is the structural schematic diagram of the focus incident determining device in one embodiment.
Specific embodiment
To facilitate the understanding of the present invention, a more comprehensive description of the invention is given in the following sections with reference to the relevant attached drawings.In attached drawing
Give preferred embodiment of the invention.But the invention can be realized in many different forms, however it is not limited to herein
Described embodiment.On the contrary, purpose of providing these embodiments is keeps the understanding to the disclosure more saturating
It is thorough comprehensive.
Unless otherwise defined, all technical and scientific terms used herein and belong to technical field of the invention
The normally understood meaning of technical staff is identical.Term used herein is only for the purpose of describing specific embodiments
It is not intended to limit the present invention.Term as used herein " and/or " it include appointing for one or more relevant listed items
Meaning and all combinations.
Fig. 1 shows the working environment schematic diagram in one embodiment of the invention, as shown in Figure 1, its working environment is related to
Terminal/server 101 and server 102, can be led between terminal/server 101 and server 102 by network
Letter.Terminal/server 101 can obtain the information for the event that various history have occurred from each server 102, and be stored in history thing
In part database, which be can be set in terminal/server 101, or is arranged in terminal/server
On database server except 101.Terminal/server 101 is obtained from server 102, can be with pass before and after the time
The sequence of events of the frequency of each event and each event of system is also possible to obtain each thing occurred from server 102
After part, the sequence of events of the frequency with time contextual each event and each event is re-formed.The present invention is implemented
Example excavates the scheme for determining focus incident what is involved is terminal/server 101.
The composed structure schematic diagram of terminal/server 101 in one embodiment is as shown in Figure 2 comprising passes through system
Processor, power supply module, storage medium, memory and the communication interface of bus connection.Wherein, the processor for provide calculate and
Control ability supports the operation of entire terminal/server.The storage medium is stored with operating system, database and a kind of hot spot
The computer applied algorithm of event determining device, the focus incident determining device realize a kind of focus incident when being executed by processor
Determine method.The interior operation for saving as the computer applied algorithm in storage medium provides environment, and communication interface is for carrying out net
Network communication.It will be understood by those skilled in the art that structure shown in Figure 2, only part relevant to example scheme is tied
The block diagram of structure does not constitute the restriction for the terminal/server being applied thereon to this embodiment scheme, specific terminal/clothes
Business device may include perhaps combining certain components or with different component cloth than more or fewer components as shown in the figure
It sets.
Fig. 3 shows the flow diagram that the focus incident in one embodiment determines method, as shown in figure 3, the implementation
Focus incident in example determines that method includes step S301 to step S304.
Step S301: obtaining each sequence of events in history event data library, any one sequence of events includes when having
Between the frequency of contextual each event and each event.
One sequence of events constitutes a complete information flow, includes more than two in a sequence of events
Event, each event can also be referred to as project in a particular application, and an event package contains specific event title, while can be with
Timestamp mark is carried either to the mark (such as event number) for indicating chronological order, to identify an event sequence
The context between each event in column.By taking event number as an example, can be set as the bigger event of event number event number more
Occur after small event.
In history event data library, it can wrap containing multiple sequences of events, each sequence of events can also be sent out based on event
Raw sequencing is ranked up.It, can be based on the generation of first event in each sequence of events in a specific example
Time determines the sequencing of each sequence of events.
Step S302: to any one sequence of events, according to the frequency of each event in the sequence of events, determining should
Event value parameter value of each event in the sequence of events in sequence of events；Also according to the sequence of events in the historical events
Sequence order in database determines the recent validity parameter value of the sequence of the sequence of events, and according to the sequence of the sequence of events
Recent validity parameter value is arranged, determines the recent validity parameter of event of each event in the sequence of events in the sequence of events
Value.
Here event value parameter value, embodies the value or risk of event to a certain extent.In one example,
Determine that the mode of event value parameter value of the event in any one sequence of events in the sequence of events may is that basis should
The event unit of value value of each event in sequence of events and hair of each event in the sequence of events in the sequence of events
Raw number, determines event value parameter value of each event in the sequence of events.Wherein, event unit of value value can be in conjunction with real
Border needs customized setting.In an application example, it can be the event unit of value value of event and the event in the thing
The product of frequency in part sequence, as event value parameter value of the event in the sequence of events.
Here the recent validity parameter value of the sequence of sequence of events, embodies the recent of sequence of events to a certain extent
Effectiveness.When determining the recent validity parameter value of sequence of sequence of events in one example, can according to time decay factor,
The sequence order of sequence of events number and the sequence of events in the history event data library in history event data library is true
It is fixed, to obtain recent virtual value related to time.The time decay factor can be a customized numerical value.
It, can be by the recent validity parameter of the sequence after the recent validity parameter value of sequence that sequence of events has been determined
Value, as the recent validity parameter value of event of each event in the sequence of events in the sequence of events.
Step S303: any one event is determined according to event value parameter value of the event in each sequence of events
The event value parameter total value of the event, and according to the recent validity parameter value of event of the event in each sequence of events, really
The recent validity total value of event of the fixed event.
In one example, it in the event value parameter total value for the event that determines, for any one event, can be
By event value parameter value of the event in each sequence of events and value, event value parameter total value as the event.?
When determining the recent validity total value of the event of event, for any one event, it can be the event in each sequence of events
The middle recent validity parameter value of event and value, the recent validity total value of event as the event.
Step S304: according to the recent validity total value of the event value parameter total value and event of each event, from each event
Determine focus incident.
In a specific example, event value threshold can be more than or equal in the event value parameter total value of event
When the recent validity total value of event of value and the event is more than or equal to recent virtual value threshold value, determine that the event is hot spot
Event.
Based on the scheme of embodiment as described above, based on contextual with the time in history event data library
The sequence of events of the frequency of each event and each event, by the event value parameter for determining each event in sequence of events
Value, and determine therefrom that the event value parameter total value of each event, and the recent validity ginseng of the sequence by determining each sequence of events
Numerical value determines therefrom that the recent validity total value of the event of each event, thus event value parameter total value and thing based on each event
The recent validity total value of part determines focus incident, thus by being carried out based on the sequence of events of temporal aspect to focus incident
It excavates and determines, breach the limitation that not can be carried out the focus incident for containing temporal aspect prediction, improve focus incident prediction
Accuracy and performance.
It may include identical two for two or more sequences of events in some technology scenes
A above event, and tandem of the event more than the two in these sequences of events is identical, in the case, for
The purpose for reducing calculation amount, can account for as a whole for the event more than the two.In one embodiment
In, the subsequence integrally formed that can form the event more than the two includes the subsequence of at least two event, claims
Be an event.At this point, being referred to as subevent for the event of the script in the event (i.e. the subsequence), i.e., in this situation
Under, this kind of event is to contain at least two the subsequence of subevent.Wherein, the event value parameter of this kind of event (i.e. subsequence)
Value be the subsequence in each subevent event value parameter value and value.
In one embodiment, which can be the sequence of events in history event data library, comprising the subsequence
Number be more than or equal to minimum support threshold subsequence.I.e. in the whole thing formed comprising more than two events
It just include the entirety of more than two events as one using this when the number of part sequence is greater than or equal to minimum support threshold
A subsequence (event of an entirety in other words) is treated.
It in the case, in one example, can be by the sub- sequence when the focus incident of above-mentioned determination is subsequence
Time posterior subevent is determined as the focus incident that will occur of prediction, i.e., time posterior sub- thing in the subsequence in column
Part is predicted to be the focus incident that will occur, to realize the prediction to following focus incident simultaneously.
It, can also be into after each sequence of events in above-mentioned acquisition history event data library in an application example
One step generates sequence of events tree, reverse indexing table and look-up table.It then, can be according to each event sequence in history event data library
The sequence order of column executes following processes to any one sequence of events:
Using event as node, the information of each event in each sequence of events is stored in the sequence tree, the letter of event
Breath includes: child node, the event of node where the father node of node, event where the nodename of node, event where event
Event value parameter value and the recent validity parameter value of event；
The mark of sequence of events and the index information of each event in the sequence of events are added in reverse indexing table；
The mark of sequence of events is added in a lookup table, and will be each in the mark direction of the sequence of events sequence of events
The last one node of event in sequence tree.
Thus, can be to store sequence of events based on the storage organization of sequence tree, reverse indexing table and look-up table
Crucial useful information.Wherein, reverse indexing table can be used to rapidly find out the correlated series that given event occurs, and look-up table
To link sequence tree and reverse indexing table, it is directed toward the last one tree node of each sequence of events in sequence tree, makees
The set of dependent event sequence rapidly can be rapidly retrieved from sequence tree with being to identify by sequence of events
It in one example, can be using the subsequence as one for the above-mentioned subsequence comprising more than two subevents
A tree node is added in sequence tree.At this point, before each sequence of events is added in sequence tree, can to each sequence of events into
Row analysis, when the number (i.e. the number of the sequence of events comprising the subsequence) for the sequence of events for the place of some subsequence occur
It, can be using the subsequence as a new item (new in other words event, subsequence event when more than or equal to minimum support threshold
Deng), it is added in sequence tree as a node.So as to reduce the number of nodes of sequence tree accordingly, with compressed sequence tree
Memory space.
Wherein, in one example, time preceding first event is not existing sequence tree in sequence of events
When root node, the root node of first event is created in sequence tree.
In one example, can also include the following steps:
It identifies the branch of only one leaf node in the sequence tree, and the branch of only one each leaf node is divided
A tree node He Bing be expressed as.
To be directed to the branch of only one leaf node, a tree node can be merged into and carried out in sequence tree
It indicates, so as to reduce the number of nodes of sequence tree, the memory space of compressed sequence tree.
Based on embodiment as described above, specifically it is illustrated using example below in conjunction with one of them.
The temporal aspect for being related to timeliness, which becomes in network event development, needs an important feature to be applied, such as daily
The focus incident of internet outburst in life, the under cover extremely important useful value information of successive order of occurrence.Big portion
Divided data belongs to series model, themselves has certain continuity, therefore a model that can portray temporal aspect is to realize
The important foundation that data content understands.Moreover, for different focus incidents, in the value (value, risk that contain
Value, weight, interest-degree etc.) different, of the invention specific example, the temporal aspect occurred in conjunction with historical events develops and
Gesture and frequency characteristic detect and determine focus incident, and the network hotspot event that predictable future may occur on this basis, into
And it can achieve the purpose of information recommendation and usertracking.
In the concrete application example, it is necessary first to obtain each sequence of events from history event data library, the history thing
Part database is the database of a timing, and the data stored in database have timing.In one concrete application example
History event data library storage content can be as shown in table 1 below.In shown in the following table 1, thus it is shown that original storage in database
Content example.For the purpose of brief description, 5 sequences of events in database are illustrated only in table 1.
Table 1: the history event data library (original) of timing type
Event (project) " Shenzhen landslide " in table 1 is indicated with digital code A, " the first survivor is rescued " uses word
Female code name B indicates, " victim stand in silent tribute activity " is indicated with digital code C, " deep case-involving suspect " is indicated with digital code D, " the
Two World Wide Web conferences " indicate that then table 1 can be exchanged into shown in the following table 2 with digital code A.
Table 2
| Sequence | Event (project): frequency |
| s1 | {A}:3,{B}:6,{C}:2 |
| s2 | {A}:1,{B}:3 |
| s3 | {A}:9,{B}:3,{D}:8,{C}:5 |
| s4 | {B}:4,{C}:2 |
| s5 | {E}:3,{A}:2,{B}:3,{A}:1 |
In upper table 1, table 2, sequence indicates a sequence of events, and a sequence of events is a complete information flow.Wherein,
S1, s2, s3, s4, s5 are the mark of each sequence of events, and can be by certain rule compositor, example between s1, s2, s3, s4, s5
As according to time sequence (sequence of the generation event such as based on first event in s1, s2, s3, s4, s5).
Above-mentioned event (project) is minimum composition unit in sequence of events, for a sequence of events, by multiple
Event composition, and this multiple event is ordered into, as event (project) in the sequence of events s1 in table 2 includes A, B,
C}.Wherein, event can be stabbed with perhaps binding time to identify or can be the mark to indicate chronological order
(such as event number eventid), to identify the context between each event in a sequence of events.With event number
For eventid, it is believed that event number eventid bigger event occurs after the smaller event of eventid.Above-mentioned
It also include the information of the frequency of each event in table 2, in the sequence of events s1 in table 2, event A has occurred 3 times, thing
Part B has occurred 6 times, and event C has occurred 2 times.
Based on above content it was determined that the history event data library in this example includes multiple sequences (i.e. event sequence
Column), each sequence of events is made of multiple event event, and multiple events in each sequence of events are that the time is orderly, because
This, this exemplary history event data library is sequence database.
After obtaining each sequence of events in history event data library, that is, it can determine that each event in each sequence of events exists
Event value parameter value in each sequence of events, i.e., the value of each event in each sequence of events.In order to determine in sequence of events
Each event in event value parameter value, need first to obtain the unit value (event unit of value value) of each event.In conjunction with table 2
Shown, the unit value (event unit of value value) of each event in an application example is as shown in table 3 below:
Table 3: the unit value table of event
| Event | A | B | C | D | E |
| Unit value | 6 | 12 | 1 | 9 | 3 |
Shown in table 3 as above, the corresponding unit value of each event in the history event data library of timing is had recorded in the table
(event unit of value value), unrelated with sequence of events, the unit value of an event is after determination, although occurring from not
With sequence of events in, unit value will not change, thus can also be referred to as unit external value, be value outside or
A kind of form of expression of risk, expression is that the corresponding unit value (or risk) of a certain part event occurs, as shown in Figure 3
In, it is 6 that primary unit value, which occurs, for event A；It is 12 that primary unit value, which occurs, for event B；Event C, D, E occur primary
Unit value is respectively 1,9,3.
Event in unit value (event unit of value value) based on each event and current event sequence is in the event
Frequency in sequence, it may be determined that event value parameter value of the event in the sequence of events, the event value parameter value
It is specifically as follows the event value parameter value of the event and the product of frequency of the event in the sequence of events.In conjunction with table
2, shown in table 3, event A is in the event value parameter value (value of the event A in sequence of events s1) in sequence of events s1 are as follows: 6
(unit value of event A) × 3 (frequency=18 of the event A in sequence s1).Other each events are in sequence of events
Event value parameter value can refer to same mode and handled.
In one example, each event is being obtained after the event value parameter value (value) in each sequence of events, it can
The further event value parameter value according to each event in sequence of events determines the sequence value parameter value of each sequence of events,
Then according to the sequence value parameter value of each sequence of events, the value parameter total value in the history event data library is determined.
Wherein, in the event value parameter value according to each event in the sequence of events, the sequence of the sequence of events is determined
When value parameter value, can by event value parameter value of each event in the sequence of events in the sequence of events and value,
Sequence value parameter value as the sequence of events.Sequence value parameter value embodies the value or risk of sequence of events.Knot
Close table 2, shown in table 3, sequence of events s2 includes event A and B, sequence of events s5 includes event E, A and B, then sequence of events s2
Sequence value parameter value (value of sequence s2) are as follows: value (the event value parameter value) 6 × 1+ event of event A in sequence s2
Value (the event value parameter value) 12 × 3=42 of B in sequence s2.Similarly, the sequence value parameter value of sequence of events s5
(value of sequence s5) is 3 × 3+2 × 6+3 × 12+1 × 6=63.
And in the event value parameter value according to each sequence of events, determine the value parameter total value in the history event data library
When, can by the event value parameter value of each sequence of events in history event data library and value, as history event data
The value parameter total value in library.In conjunction with shown in table 2, table 3, the value parameter total value (total value) of the database comprising sequence s1 to s5
Are as follows: 92+42+167+50+63=414.
On the other hand, in one example, event value parameter value (valence of each event in each sequence of events is being obtained
Value) after, it can determine the event value parameter total value for obtaining each event.For any one event, it can be this
Event value parameter value of the event in each sequence of events and value, the event value parameter total value as the event.In conjunction with table 2
It is shown, then occur in sequence of events s1, s3, s4 due to event C, then the event value parameter total value of event C are as follows: event C exists
Event value parameter value+event C in sequence s1 is in event value parameter value+thing of the event C in sequence s4 in sequence s3
Part value parameter value.
It is then determined that the recent validity parameter value of the sequence of each sequence of events, the recent validity parameter value of the sequence is one
Determine the recent effectiveness that part sequence of getting over is embodied in degree.The recent validity parameter value of sequence of events is determined in one example
When, can according in time decay factor, history event data library sequence of events number and the sequence of events in the history
Sequence order in event database determines, so that virtual value related to time is obtained, to indicate the close of the sequence of events
Phase validity.The time decay factor can be a customized numerical value.
Determine that the recent validity parameter value of sequence of sequence of events can use following formula in one concrete application example
It carries out:
Wherein, ScurrentIndicate the sequence of events number in history event data library, SqIndicate the sequence of events in the history
Sequence order in event database, δ indicate customized time decay factor, value interval be δ ∈ (0,1].
Assuming that 0.1 is set by δ, and history event data library shown in table 2 includes 10 sequences of events in total, that
, the recent validity parameter value of the sequence of sequence of events s1 is R (s1)=(1-0.1)(10-1)=0.3874, sequence of events s2's
The recent validity parameter value of sequence is R (s2)=(1-0.1)(10-2)=0.4305.
After the recent validity parameter value of sequence that sequence of events has been determined, the sequence of the sequence of events can be had in the recent period
Effect property parameter value, as the recent validity parameter value of event of each event in the sequence of events in the sequence of events.
Determining each event after the recent validity parameter value of event of each sequence of events, it may be determined that the thing of each event
The recent validity total value of part, the recent validity total value of the event specifically can be event event in each sequence of events to be had in the recent period
Effect property parameter value and value.
After the recent validity total value of event value parameter total value and event for determining each event, that is, it can determine the event
It whether is focus incident.In one example, from each event determine focus incident when, can event event value parameter it is total
Value is more than or equal to event value threshold value (alternatively referred to as minimum value threshold value) and the recent validity total value of event of the event
When more than or equal to recent virtual value threshold value when (alternatively referred to as minimum effective threshold value in the recent period), determine that the event is hot spot thing
Part.I.e. when the event value parameter total value of event is effective in the recent period more than or equal to the event of event value threshold value and the event
Property total value when being more than or equal to recent virtual value threshold value, which is referred to alternatively as effective high value event in the recent period, and can will
It is determined as it being focus incident.Event value threshold value, recent virtual value threshold value can carry out customized setting.
In the implementation procedure of above scheme, it can be further optimized, for the frequency in history event data library
The Frequent Subsequence can be integrally used as an event by numerous subsequence, complete the above process.
In the scheme of the present embodiment, for sequence x and sequences y, if there is the mapping of a sequence, so that in x
Each event be contained in some event in y, then referred to as x is contained in y, and x is the subsequence of y, such as sequence D -> A
It is sequence D -> AB- > A subsequence.
The number of sequence in entire sequence sets comprising sequence s is denoted as to the support of sequence s, then if there is a sequence
S, it is the subsequence of certain sequence of data set D, then its support are as follows: n sequence of support (the s)=sequence s in data set D
The number occurred in column.
In the case, by the way that minimum support threshold min_support is arranged, then when the support of some sequence s
When support (s) is more than or equal to minimum support threshold min_support, sequence s can be known as Frequent episodes,
Or referred to as sequence pattern.The effect of sequence pattern is, it is assumed that has obtained < { Shenzhen landslide }, { the first survivor obtains
Rescue } > as sequence pattern (angle brackets indicate what the element inside it was ordered into, previous to occur before the latter), that
It can think, in case of event { Shenzhen landslide }, then event very likely occurs later, and { the first survivor obtains
It rescues }.
It accordingly,, can be to history after obtaining the sequence of events in history event data library in the specific example scheme
Event data is analyzed, and determines the subsequence for reaching minimum support threshold in database with the presence or absence of support, and will
The subsequence is integrally analyzed as an event.
Wherein, for subsequence, value (event value ginseng of the subsequence in some sequence of events belonging to it
Numerical value) are as follows: value of each event in the sequence of events in the subsequence and value, should and value can be adduction.In conjunction with table
It 2, only include value of the subsequence BC of event B, C in sequence of events s3 are as follows: value of the event B in sequence s3 shown in table 3
Value 1 × 5=41 of 12 × 3+ event C in sequence s3.
After the value in each sequence of events for obtaining subsequence belonging to it (or including it), to the subsequence
Value summation in all sequences of events comprising it, to obtain the total value of the subsequence in the database, the i.e. son
The event value parameter total value of sequence.
On the other hand, for subsequence, the subsequence having in the sequence of events comprising it in the recent period can also be calculated
Valid value (i.e. the recent validity parameter value of event of the subsequence in the sequence of events), in specific example, subsequence is at this
Recent virtual value in sequence of events, can be equal to the recent virtual value of the sequence of events, (i.e. the sequence of the sequence of events has in the recent period
Effect property parameter value).Summation is added up to recent virtual value of the subsequence in each sequence of events that it occurs, it is hereby achieved that
The recent virtual value (i.e. the recent validity total value of the event of the subsequence) of the subsequence in the database.
Determining each subsequence total value (i.e. the event value parameter total value of subsequence) in the database and in data
After recent virtual value (i.e. the recent validity total value of the event of subsequence) in library, which can be combined other times
Analyze focus incident.In one example, event value threshold can be more than or equal in the total value of subsequence in the database
It is worth (alternatively referred to as minimum value threshold value) and the recent virtual value of the subsequence in the database is more than or equal in the recent period effectively
When being worth threshold value (alternatively referred to as minimum effective threshold value in the recent period), determines that the event is focus incident, determine that the subsequence is to have in the recent period
The high value subsequence of effect.The posterior subevent of time in subsequence can also be determined as to the heat that will occur of prediction simultaneously
Point event, i.e., the posterior subevent of event is predicted to be the focus incident that will occur in the subsequence, to realize simultaneously
Prediction to following focus incident.Event value threshold value, recent virtual value threshold value can carry out customized setting.
In a specific example, sequence of events tree, reverse indexing table and look-up table can also be further generated.With
Afterwards, following mistakes can be executed to any one sequence of events according to the sequence order of each sequence of events in history event data library
Journey:
Using event as node, the information of each event in each sequence of events is stored in the sequence tree, the letter of event
Breath includes: child node, the event of node where the father node of node, event where the nodename of node, event where event
Event value parameter value and the recent validity parameter value of event；
The mark of sequence of events and the index information of each event in the sequence of events are added in reverse indexing table；
The mark of sequence of events is added in a lookup table, and will be each in the mark direction of the sequence of events sequence of events
The last one node of event in sequence tree.
For the schematic diagram in conjunction with shown in table 2, table 3, it is assumed that the practical corresponding database of table 2 includes 10 sequences of events,
Be inserted into first sequence of events s1<A, B, C>after sequence tree, reverse indexing table and look-up table schematic diagram such as Fig. 4 institute
Show, shown in Fig. 4 in, A:18,0.3874 indicates value (event value parameter value) of the event A in sequence s1, and 0.3874 indicates
The recent virtual value of sequence s1 namely the recent virtual value of the event A after insetion sequence s1.
And then, be inserted into second sequence of events s2<A, B>after sequence tree, reverse indexing table and look-up table signal
Figure is as shown in figure 5, in shown in Fig. 5, and the value of node A, B and recent virtual value are due to the number that has been superimposed in time series s2
According to be changed.Similarly, be inserted into third sequence of events s3<A, B, D, C>after sequence tree, reverse indexing table
And look-up table schematic diagram as shown in fig. 6, insertion the 4th sequence of events s4<B, C>after sequence tree, reverse indexing table and
The schematic diagram of look-up table is as shown in Figure 7.As shown in fig. 7, in insertion s4<B, C>when, since the node of already present event B is not
The root node of sequence tree, to need to create the root node of an event B in sequence tree.Similarly, it is inserted into the 5th event
Sequence s5<E, A, B, A>after sequence tree, the schematic diagram of reverse indexing table and look-up table it is as shown in Figure 8.
With reference to the sequence tree shown in fig. 8 initially set up, it can be found that sequence of events s1, s2, s3, s5 has in sequence tree
There is common subsequence<A, B>, it is assumed that minimum support threshold is 3, which is Frequent Subsequence, it is assumed that minimum value threshold
Value is 150, then the subsequence is high value subsequence.In this case, it is possible to the subsequence<A, B>compressed, with section
Save the memory space of sequence tree.
In a wherein example, subsequence<A, B>be can identified based on the sequence of events analysis in database
When Frequent Subsequence, then corresponding new item is generated in set, and when being inserted into each sequence of events, be related to the sub- sequence
Column<A, B>when, all substituted with the corresponding new item.Assuming that by the subsequence<A, B>substituted with new item X then has compressed frequency
Numerous subsequence<A, B>after sequence tree schematic diagram it is as shown in Figure 9.Comparison diagram 8 and Fig. 9 as it can be seen that script initial construction sequence
Tree needs 11 nodes, but generates by above-mentioned for Frequent Subsequence after new item comes instead, it is only necessary to 9 nodes, to save
The about memory space of sequence tree.
On the other hand, with reference to the sequence tree shown in fig. 8 initially set up, it is found that in the sequence tree being initially generated
It include the branch of only one leaf node, the branch of only one this leaf node can occupy more node storage
Space therefore in one example, further this kind of branch can also be merged, i.e., by only one leaf node
Branch merge be expressed as a tree node, to reduce the number of nodes of sequence tree.
For the sequence tree shown in fig. 8 initially set up, only one leaf node in sequence tree is being identified
After subsequence branch, it is assumed that subsequence branch<D, C>merging postscript be DC, merge subsequence branch<D, C>after sequence tree
Schematic diagram it is as shown in Figure 10.Likewise, it is assumed that subsequence branch<B, C>merging postscript be BC, merge subsequence branch<B, C>
The schematic diagram of sequence tree later is as shown in figure 11.Assuming that subsequence branch<E, A, B, A>merging postscript is EABA, merges sub- sequence
Arrange branch<E, A, B, A>after sequence tree schematic diagram it is as shown in figure 12.
Comparison diagram 8 and Figure 12 are as it can be seen that the sequence tree of script initial construction needs 11 nodes, but passes through above-mentioned merging 3
After subsequence branch, it is only necessary to 6 nodes, to save the memory space of sequence tree.
Wherein, in another example, above-mentioned compression Frequent Subsequence, generate for Frequent Subsequence the compression side of new item
Formula, and for the compress mode that merges of subsequence branch of only one leaf node, can carry out simultaneously.For figure
The sequence tree initially set up shown in 8 is applied for one and compresses Frequent Subsequence in example and merge the sequence after subsequence branch
The schematic diagram for arranging tree is as shown in figure 13, comparison diagram 8 and Figure 13 as it can be seen that the sequence tree of script initial construction needs 11 nodes, but
After being compressed simultaneously using above two compress mode, it is only necessary to which 5 nodes have farthest saved sequence tree
Memory space.
Based on thought same as mentioned above, a kind of focus incident determining device is also provided in an example.Figure 14 shows
The structural schematic diagram of the focus incident determining device in an example is gone out.As shown in figure 14, the focus incident in the example is true
Determining device includes: that acquisition module 1401, value determining module 1402, recent validity determining module 1403 and focus incident are true
Cover half block 1404.
Wherein, module 1401, for obtaining each sequence of events in history event data library, any one event sequence are obtained
Column include the frequency with time contextual each event and each event；
One sequence of events constitutes a complete information flow, includes more than two in a sequence of events
Event, each event can also be referred to as project in a particular application, and an event package contains specific event title, while can be with
Timestamp mark is carried either to the mark (such as event number) for indicating chronological order, to identify an event sequence
The context between each event in column.By taking event number as an example, can be set as the bigger event of event number event number more
Occur after small event.
In history event data library, it can wrap containing multiple sequences of events, each sequence of events can also be sent out based on event
Raw sequencing is ranked up.It, can be based on the generation of first event in each sequence of events in a specific example
Time determines the sequencing of each sequence of events.
It is worth determining module 1402, is used for any one sequence of events, according to the hair of each event in the sequence of events
Raw number, determines event value parameter value of each event in the sequence of events in the sequence of events；It is also used to any one
A event determines the event value parameter total value of the event according to event value parameter value of the event in each sequence of events.
Here event value parameter value, embodies the value or risk of event to a certain extent.In one example,
Determine that the mode of the event value parameter value of the event in any one sequence of events may is that according to each in the sequence of events
The event unit of value value of event and frequency of each event in the sequence of events in the sequence of events determine each
Event value parameter value of the event in the sequence of events.Wherein, event unit of value value can be customized in conjunction with being actually needed
Setting.In an application example, it can be the event unit of value value of event and hair of the event in the sequence of events
The product of raw number, as event value parameter value of the event in the sequence of events.
Here the recent validity parameter value of sequence of events embodies the recent effective of sequence of events to a certain extent
Degree.It, can be according to time decay factor, historical events when determining the recent validity parameter value of sequence of events in one example
The sequence order of sequence of events number and the sequence of events in the history event data library in database is determining, thus
Obtain virtual value related to time.The time decay factor can be a customized numerical value.
It, can be by the recent validity parameter value, as this after the recent validity parameter value of sequence of events has been determined
The recent validity parameter value of event of each event in the sequence of events in sequence of events.
Recent validity determining module 1403, is used for any one sequence of events, according to the sequence of events in the history
Sequence order in event database determines the recent validity parameter value of the sequence of events, and according to the sequence of the sequence of events
Recent validity parameter value is arranged, determines the recent validity parameter of event of each event in the sequence of events in the sequence of events
Value；It is also used to any one event, according to the recent validity parameter value of event of the event in each sequence of events, determining should
The recent validity total value of the event of event.
In one example, it in the event value parameter total value for the event that determines, for any one event, can be
By event value parameter value of the event in each sequence of events and value, event value parameter total value as the event.?
When determining the recent validity total value of the event of event, for any one event, it can be the event in each sequence of events
The middle recent validity parameter value of event and value, the recent validity total value of event as the event.
Focus incident determining module 1404, for the recent validity of event value parameter total value and event according to each event
Total value determines focus incident from each event.
In a specific example, event value threshold can be more than or equal in the event value parameter total value of event
When the recent validity total value of event of value and the event is more than or equal to recent virtual value threshold value, determine that the event is hot spot
Event.
It may include identical two for two or more sequences of events in some technology scenes
A above event, and tandem of the event more than the two in these sequences of events is identical, in the case, for
The purpose for reducing calculation amount, can account for as a whole for the event more than the two.In one embodiment
In, the subsequence integrally formed that can form the event more than the two includes the subsequence of at least two event, claims
Be an event.At this point, being referred to as subevent for the event of the script in the event (i.e. the subsequence), i.e., in this situation
Under, this kind of event is to contain at least two the subsequence of subevent.Wherein, the event value parameter of this kind of event (i.e. subsequence)
Value be the subsequence in each subevent event value parameter value and value.
In one embodiment, which can be the sequence of events in history event data library, comprising the subsequence
Number be more than or equal to minimum support threshold subsequence.I.e. in the whole thing formed comprising more than two events
It just include the entirety of more than two events as one using this when the number of part sequence is greater than or equal to minimum support threshold
A subsequence (event of an entirety in other words) is treated.
In the case, in one example, focus incident determining module 1404 is subsequence in determining focus incident
When, time posterior subevent in the subsequence can be determined as to the focus incident that will occur of prediction, the i.e. sub- sequence
The time, posterior subevent was predicted to be the focus incident that will occur in column, to realized simultaneously to following focus incident
Prediction.
In one example, as shown in figure 14, which can also include:
Sequence tree processing module 1405, for generating sequence of events tree, and according to each event in history event data library
The sequence order of sequence, to any one sequence of events, using event as node, by the information of each event in each sequence of events
Be stored in the sequence tree, the information of event include: the father node of node where the nodename of node, event where event,
Child node, the event value parameter value of event and the recent validity parameter value of event of node where event；
Reverse indexing table handing module 1406, for generating reverse indexing table, and according to each in history event data library
The sequence order of sequence of events is added the mark of sequence of events and is somebody's turn to do in reverse indexing table to any one sequence of events
The index information of each event in sequence of events；
Table handing module 1407 is searched, for generating look-up table, and according to each sequence of events in history event data library
Sequence order the mark of sequence of events added in a lookup table to any one sequence of events, and by the mark of the sequence of events
Know the last one node of each event being directed toward in the sequence of events in sequence tree.
Thus, can be to store sequence of events based on the storage organization of sequence tree, reverse indexing table and look-up table
Crucial useful information.Wherein, reverse indexing table can be used to rapidly find out the correlated series that given event occurs, and look-up table
To link sequence tree and reverse indexing table, it is directed toward the last one tree node of each sequence of events in sequence tree, makees
The set of dependent event sequence rapidly can be rapidly retrieved from sequence tree with being to identify by sequence of events.
In one example, the sequence tree processing module 1405, time preceding first event also in sequence of events
When not being the root node of existing sequence tree, the root node of first event is created in sequence tree.
As shown in figure 14, the focus incident determining device in an example can also include:
Branch's merging module 1408, for identification in the sequence tree only one leaf node branch, and by each
There is the branch of a leaf node to merge respectively and is expressed as a tree node.
Based on example as described above, a kind of computer equipment is also provided in one embodiment, the computer equipment packet
The computer program that includes memory, processor and storage on a memory and can run on a processor, wherein processor executes
Realize that any one focus incident in the various embodiments described above such as determines method when described program.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, it is non-volatile computer-readable that the program can be stored in one
It takes in storage medium, in the embodiment of the present invention, which be can be stored in the storage medium of computer system, and by the calculating
At least one processor in machine system executes, and includes the process such as the embodiment of above-mentioned each method with realization.Wherein, described
Storage medium can be magnetic disk, CD, read-only memory (Read-Only Memory, ROM) or random access memory
(Random Access Memory, RAM) etc..
Accordingly, a kind of storage medium is also provided in one embodiment, is stored thereon with computer program, wherein the journey
Realize that any one focus incident in the various embodiments described above such as determines method when sequence is executed by processor.
Each technical characteristic of embodiment described above can be combined arbitrarily, for simplicity of description, not to above-mentioned reality
It applies all possible combination of each technical characteristic in example to be all described, as long as however, the combination of these technical characteristics is not deposited
In contradiction, all should be considered as described in this specification.
The embodiments described above only express several embodiments of the present invention, and the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to protection of the invention
Range.Therefore, the scope of protection of the patent of the invention shall be subject to the appended claims.
Claims (10)
1. a kind of focus incident determines method, which is characterized in that comprising steps of
Each sequence of events in history event data library is obtained, any one sequence of events includes having the time contextual each
The frequency of event and each event；
Any one sequence of events is determined in the sequence of events according to the frequency of each event in the sequence of events
Event value parameter value of each event in the sequence of events；Also according to sequence of the sequence of events in the history event data library
Column sequence, determines the recent validity parameter value of the sequence of the sequence of events, and according to the recent validity of the sequence of the sequence of events
Parameter value determines the recent validity parameter value of event of each event in the sequence of events in the sequence of events；
The event of the event is determined according to event value parameter value of the event in each sequence of events to any one event
Value parameter total value, and according to the recent validity parameter value of event of the event in each sequence of events, determine the thing of the event
The recent validity total value of part；
According to the recent validity total value of the event value parameter total value and event of each event, focus incident is determined from each event.
2. focus incident according to claim 1 determines method, which is characterized in that including at least one in following two
:
To any one sequence of events, according to the event unit of value value of each event in the sequence of events and the event sequence
Frequency of each event in the sequence of events in column determines event value parameter of each event in the sequence of events
Value；
To any one sequence of events, according to the sequence of events number in time decay factor, history event data library and it is somebody's turn to do
Sequence order of the sequence of events in the history event data library determines the recent validity parameter value of the sequence of the sequence of events.
3. focus incident according to claim 1 determines method, which is characterized in that event is to include at least one subevent
Subsequence, the focus incident be subsequence when, by time posterior subevent in the subsequence be determined as prediction
The focus incident that will occur.
4. determining method to focus incident described in 3 any one according to claim 1, which is characterized in that obtaining history thing
After each sequence of events in part database, further comprise the steps of:
Generate sequence of events tree, reverse indexing table and look-up table；According to the sequence of each sequence of events in history event data library
Column sequence, to any one sequence of events, executes following processes:
Using event as node, the information of each event in each sequence of events is stored in the sequence tree, the packet of event
It includes: child node, the thing of event of node where the father node of node, event where the nodename of node, event where event
Part value parameter value and the recent validity parameter value of event；
The mark of sequence of events and the index information of each event in the sequence of events are added in reverse indexing table；
The mark of sequence of events is added in a lookup table, and the mark of the sequence of events is directed toward to each event in the sequence of events
The last one node in sequence tree.
5. focus incident according to claim 4 determines method, which is characterized in that further comprise the steps of: the identification sequence
The branch of only one leaf node in tree, and the branch of only one each leaf node is merged respectively and is expressed as a burl
Point.
6. a kind of focus incident determining device characterized by comprising
Module is obtained, for obtaining each sequence of events in history event data library, any one sequence of events includes when having
Between the frequency of contextual each event and each event；
It is worth determining module, is used for any one sequence of events, according to the frequency of each event in the sequence of events, really
Event value parameter value of each event in the sequence of events in the fixed sequence of events；It is also used to any one event, root
According to event value parameter value of the event in each sequence of events, the event value parameter total value of the event is determined；
Recent validity determining module, is used for any one sequence of events, according to the sequence of events in the history event data
Sequence order in library, determines the recent validity parameter value of the sequence of events, and is had in the recent period according to the sequence of the sequence of events
Effect property parameter value, determines the recent validity parameter value of event of each event in the sequence of events in the sequence of events；Also use
In determining the event according to the recent validity parameter value of event of the event in each sequence of events to any one event
The recent validity total value of event；
Focus incident determining module, for the recent validity total value of event value parameter total value and event according to each event, from
Focus incident is determined in each event.
7. focus incident determining device according to claim 6, which is characterized in that event is to include at least one subevent
Subsequence；
The focus incident determining module is when the focus incident is subsequence, by the posterior sub- thing of time in the subsequence
Part is determined as the focus incident that will occur of prediction.
8. focus incident determining device according to claim 6 or 7, which is characterized in that further include:
Sequence tree processing module, for generating sequence of events tree, and according to the sequence of each sequence of events in history event data library
Column sequence, to any one sequence of events, using event as node, is stored in institute for the information of each event in each sequence of events
Sequence tree is stated, the information of event includes: the father node of node where the nodename of node, event where event, event place
The child node of node, the event value parameter value of event and the recent validity parameter value of event；
Reverse indexing table handing module, for generating reverse indexing table, and according to each sequence of events in history event data library
Sequence order the mark and the sequence of events of sequence of events are added in reverse indexing table to any one sequence of events
In each event index information；
Table handing module is searched, for generating look-up table, and it is suitable according to the sequence of each sequence of events in history event data library
Sequence adds the mark of sequence of events, and the mark of the sequence of events is directed toward should to any one sequence of events in a lookup table
The last one node of each event in sequence tree in sequence of events.
9. a kind of computer equipment, including memory, processor and it is stored on the memory and can be in the processor
The computer program of upper operation, which is characterized in that the processor realized when executing the computer program as claim 1 to
The step of method described in 5 any one.
10. a kind of computer storage medium, is stored thereon with computer program, which is characterized in that the program is executed by processor
The step of method of the Shi Shixian as described in claim 1 to 5 any one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710629638.5A CN109766367B (en) | 2017-07-28 | 2017-07-28 | Hot event determination method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710629638.5A CN109766367B (en) | 2017-07-28 | 2017-07-28 | Hot event determination method and device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109766367A true CN109766367A (en) | 2019-05-17 |
| CN109766367B CN109766367B (en) | 2021-06-08 |
Family
ID=66448657
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710629638.5A Active CN109766367B (en) | 2017-07-28 | 2017-07-28 | Hot event determination method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109766367B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO1991002316A1 (en) * | 1989-07-28 | 1991-02-21 | Udo Quilitz | Process for allocating data sets to time values of a consecutive sequence |
| CN101819573A (en) * | 2009-09-15 | 2010-09-01 | 电子科技大学 | Self-adaptive network public opinion identification method |
| US20110191355A1 (en) * | 2007-04-24 | 2011-08-04 | Peking University | Method for monitoring abnormal state of internet information |
| US20140280242A1 (en) * | 2013-02-25 | 2014-09-18 | Tencent Technology (Shenzhen) Company Limited | Method and apparatus for acquiring hot topics |
| CN104598539A (en) * | 2014-12-30 | 2015-05-06 | 中国联合网络通信有限公司广东省分公司 | Internet event hot degree calculation method and terminal |
| CN104598450A (en) * | 2013-10-30 | 2015-05-06 | 北大方正集团有限公司 | Popularity analysis method and system of network public opinion event |
| CN106446179A (en) * | 2016-09-28 | 2017-02-22 | 东软集团股份有限公司 | Hot topic generation method and device |
| CN106528755A (en) * | 2016-10-28 | 2017-03-22 | 东软集团股份有限公司 | Hot topic generation method and device |
| CN107066537A (en) * | 2017-03-06 | 2017-08-18 | 广州神马移动信息科技有限公司 | Hot news generation method, equipment, electronic equipment |
-
2017
- 2017-07-28 CN CN201710629638.5A patent/CN109766367B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO1991002316A1 (en) * | 1989-07-28 | 1991-02-21 | Udo Quilitz | Process for allocating data sets to time values of a consecutive sequence |
| US20110191355A1 (en) * | 2007-04-24 | 2011-08-04 | Peking University | Method for monitoring abnormal state of internet information |
| CN101819573A (en) * | 2009-09-15 | 2010-09-01 | 电子科技大学 | Self-adaptive network public opinion identification method |
| US20140280242A1 (en) * | 2013-02-25 | 2014-09-18 | Tencent Technology (Shenzhen) Company Limited | Method and apparatus for acquiring hot topics |
| CN104598450A (en) * | 2013-10-30 | 2015-05-06 | 北大方正集团有限公司 | Popularity analysis method and system of network public opinion event |
| CN104598539A (en) * | 2014-12-30 | 2015-05-06 | 中国联合网络通信有限公司广东省分公司 | Internet event hot degree calculation method and terminal |
| CN106446179A (en) * | 2016-09-28 | 2017-02-22 | 东软集团股份有限公司 | Hot topic generation method and device |
| CN106528755A (en) * | 2016-10-28 | 2017-03-22 | 东软集团股份有限公司 | Hot topic generation method and device |
| CN107066537A (en) * | 2017-03-06 | 2017-08-18 | 广州神马移动信息科技有限公司 | Hot news generation method, equipment, electronic equipment |
Non-Patent Citations (1)
| Title |
|---|
| 游丹丹: "我国网络舆情热点话题发现研究综述", 《现代情报》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109766367B (en) | 2021-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Starnini et al. | Random walks on temporal networks | |
| Yang et al. | A time efficient approach for detecting errors in big sensor data on cloud | |
| CN102770845B (en) | Optimization of archive management scheduling | |
| US9367428B2 (en) | Transparent performance inference of whole software layers and context-sensitive performance debugging | |
| CN102420831B (en) | Multi-domain network packet classification method | |
| CN109829337A (en) | A kind of method, system and the equipment of community network secret protection | |
| Rashid et al. | Dependable large scale behavioral patterns mining from sensor data using Hadoop platform | |
| Jeong et al. | Anomaly teletraffic intrusion detection systems on hadoop-based platforms: A survey of some problems and solutions | |
| CN111475838B (en) | Deep neural network-based graph data anonymizing method, device and storage medium | |
| CN110058969B (en) | Data recovery method and device | |
| CN109885456A (en) | A kind of polymorphic type event of failure prediction technique and device based on system log cluster | |
| CN110209909A (en) | Data crawling method, device, computer equipment and storage medium | |
| CN113225199A (en) | Interactive behavior prediction method and device based on time sequence network mining and electronic equipment | |
| Amiri et al. | A new efficient approach for extracting the closed episodes for workload prediction in cloud | |
| CN114430331A (en) | Network security situation sensing method and system based on knowledge graph | |
| CN110535686A (en) | Anomalous event treating method and apparatus | |
| CN109684517A (en) | A kind of historical data storage method, reading/writing method, storage device and equipment | |
| CN109766367A (en) | Focus incident determines method and device, computer equipment and storage medium | |
| Zeng et al. | An efficient index-based approach to distributed set reachability on small-world graphs | |
| CN104615697B (en) | A kind of webpage level remembers return method | |
| CN111756691A (en) | Acquisition strategy conflict detection method and device, electronic equipment and storage medium | |
| CN110221778A (en) | Processing method, system, storage medium and the electronic equipment of hotel's data | |
| CN115018124A (en) | Data prediction method, system, device and storage medium | |
| CN103440157B (en) | A kind of method and apparatus of the template for obtaining virtual machine | |
| CN103312785A (en) | Method and device for determining access relation |
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 |