CN104331493B - By the computer implemented method and device that data are explained for generating trend - Google Patents
By the computer implemented method and device that data are explained for generating trend Download PDFInfo
- Publication number
- CN104331493B CN104331493B CN201410652571.3A CN201410652571A CN104331493B CN 104331493 B CN104331493 B CN 104331493B CN 201410652571 A CN201410652571 A CN 201410652571A CN 104331493 B CN104331493 B CN 104331493B
- Authority
- CN
- China
- Prior art keywords
- search word
- search
- association
- average
- word
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
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
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 present invention provides a kind of by the computer implemented method and device that data are explained for generating trend.Methods described includes：Obtain user's search daily record；Extracted from user search daily record and one group of the first association search word of search word association search to be examined or check, correlation time point and its searching times；The first association search word, correlation time point and its searching times and user search daily record according to extraction choose searching times and change second association search word of the amplitude more than preassigned, its change direction and transformation period interval on searching times, and the change direction is for forward or backwards；Data are explained according to the second association search word and transformation period interval generation trend.In this way, automatically generating objective and degree of accuracy trend higher explains data, provide the user with objective, effective, system trend and explain, and then strengthen Consumer's Experience.
Description
Technical field
The present invention relates to microcomputer data processing, more particularly to one kind by computer implemented for generating trend
Explain the method and device of data.
Background technology
With the explosive increase of internet data, the tendency application and product launched around internet data are just continuous
Attract attention.For example, social class website can provide the Long-term change trend of user social contact active degree, ecommerce class website can be with
The Long-term change trend of user network consumption is provided, search engine class website can provide the Long-term change trend of user interest point, trend sheet
Body can be showed by the tissue to data and excavation, but how autonomous, effective, system trend be solved
It has been read as a problem.
Conventional trend interpretation method be mainly it is artificial or it is automanual trend is understood, assign subjective reason
Solution, or simple Long-term change trend is bound together with outside news.
There are the following problems for the above method：1) inclined subjectivity, due to human interpretation, different subjective bias are often presented
Different Results；2) limitation, lacks the complete analysis to possible outcome collection；3) nonautonomy, so understands and generally relies on neck
The priori of domain expert, for some conventional fields can output result, but for the arbitrary input behavior of user then without
Method obtains result；4) poor in timeliness, existing method often timely cannot provide interpretation result to described trend.
The content of the invention
It is an object of the present invention to provide it is a kind of by it is computer implemented for generate trend explain data method and
Device, is analyzed to generate the trend explanation data in the range of certain hour, so that can be automatic by searching for daily record to user
Ground generates objective and degree of accuracy trend higher and explains data.
According to an aspect of the present invention, there is provided a kind of by the computer implemented side that data are explained for generating trend
Method, including：Obtain user's search daily record；The with one group of search word association search to be examined or check is extracted from user search daily record
One association search word, correlation time point and its searching times；According to the first association search word for extracting, correlation time point and its search
Rope number of times and the user search daily record choose searching times change amplitude exceed preassigned the second association search word, its
Change direction and transformation period on searching times is interval, and the change direction is for forward or backwards；According to described second
Association search word and transformation period interval generation trend explain data.
According to another aspect of the present invention, there is provided it is a kind of for generate trend explain data device, including：Log acquisition
Unit, user obtains user's search daily record；First information extraction unit, for being treated with one group for searching for daily record extraction from described
Examine or check the first association search word, association search time point and its searching times of search word association search；Second information extraction list
Unit, for according to the first association search word, correlation time point and its searching times that extract and user search daily record choosing
Take searching times and change the second association search word, its change direction and change on searching times that amplitude exceedes preassigned
Change time interval, the change direction is for forward or backwards；Trend explains data generating unit, according to second association search
Word and transformation period interval generation trend explain data.
A kind of method and device that data are explained for generating trend provided in an embodiment of the present invention, searches for by from user
The first association search of log acquisition word, correlation time point and its searching times, then filtered out from the first association search word
Second association search word, the interval generation of change direction and transformation period according to the second association search word and its on searching times
Trend explains data, in this way, automatically generating objective and degree of accuracy trend higher explains data, provide the user with it is objective, have
Effect, the trend of system are explained, and then strengthen Consumer's Experience.
Brief description of the drawings
Fig. 1 is the method flow schematic diagram that data are explained for generating trend for showing an exemplary embodiment of the invention.
Fig. 2 is that the user's generation trend for showing another exemplary embodiment of the present invention explains that the method flow of data is illustrated
Figure.
Fig. 3 is to show the first association search word extracted from user's search daily record of exemplary embodiment of the present, close
Connection time point and its data instance figure of searching times.
Fig. 4 is the second association search word, its change direction on searching times for showing exemplary embodiment of the present
And the interval data instance figure of transformation period.
Fig. 5 is the trend matching schematic diagram of the second association search word for showing exemplary embodiment of the present.
Fig. 6 is that the transformation period Interval Maps of the second association search word for showing exemplary embodiment of the present are treated to one group
The data instance figure of the search curve of examination search word.
Fig. 7 is the data trend of the generation for showing exemplary embodiment of the present and carries out trend with reference to other data sources
The exemplary plot of explanation.
Fig. 8 is to show the related data sources exemplary plot corresponding with the data trend shown in Fig. 7.
Fig. 9 is the apparatus structure block diagram that data are explained for generating trend for showing an exemplary embodiment of the invention.
Figure 10 is the apparatus structure block diagram that data are explained for generating trend for showing another exemplary embodiment of the present invention.
Specific embodiment
Basic conception of the invention is, extracted from the search daily record of user by computer technology association search word and its
Related data content information, and the searching times based on association search word screen to association search word, by expiring for filtering out
The association search word of sufficient preassigned is interval with the transformation period of its searching times to be combined, and generation trend explains data.
Number is explained for generating trend by computer implemented to exemplary embodiment of the present below in conjunction with the accompanying drawings
According to method and device be described in detail.
Fig. 1 is the method flow schematic diagram that data are explained for generating trend for showing an exemplary embodiment of the invention.
Reference picture 1, in step S101, obtains user's search daily record, it is preferable that user's search daily record is included from many
The internet hunt record of individual data source.
In step S102, the first association search with one group of search word association search to be examined or check is extracted from user's search daily record
Word, correlation time point and its searching times.
For example, the destructuring number that the head stack that can be accessed from the target word bag, user to be examined or check or user access
According to extraction one group of search word to be examined or check.
According to the preferred embodiment of the present invention, the first association search word is extracted in the following manner：Searched from the user
Suo Zhi extract any user in a certain time interval with one group of search word to be examined or check in search word any to be examined or check
The search word together searched for (can be referred to as the search word of co-occurrence between inquiring about, for example, user A is " good by any search word to be examined or check
It is precious " scan for, after after half an hour, the user A simultaneously scans for " Heng Shi ground rice " with " Garbo ", at this point it is possible to will be " prosperous
Family name's ground rice " is used as the first association search word), and/or in any query search word with one group of search word to be examined or check
In the search word that together occurs of search word any to be examined or check (search of co-occurrence in inquiring about can be referred to as the first association search word
Word, for example, user A passes through any when examination search word " Garbo " is scanned for, the search word for simultaneously occurring also includes
" Heng Shi ground rice ", " model ice ice Li Zhiting " etc., can be using " Heng Shi ground rice ", " model ice ice Li Zhiting " etc. as the first association search
Word).
It will be appreciated by persons skilled in the art that any search word to be examined or check is likely to be chosen for the first association
Search word, now, will reject any search word to be examined or check from the result of the first association search word for selecting.
Additionally, according to a preferred embodiment of the invention, in step S102, also deleting this from the first association search word for extracting
The search word of sample：Its with one group of search word to be examined or check in the number of times that together occurs of search word any to be examined or check less than predetermined
Co-occurrence number of times so that only choose and the enough number of times of search word co-occurrence to be examined or check search word.
After the first association search word is extracted, according to predetermined timing statisticses section unit (such as 1 day), by described the
The one association search word searched statistical unit time period counts the first association search word as its correlation time point
The number of times being searched within the statistical unit time period is used as its searching times.
Fig. 3 is to show the first association search word extracted from user's search daily record of exemplary embodiment of the present, close
Connection time point and its data instance figure of searching times.
Reference picture 3, it can be seen that the search word any to be examined or check chosen from target word bag is " Garbo ", by step
After the selection of S102, the first association search word for obtaining has " Heng Shi ground rice up-to-date event ", and (correlation time point is in August, 2014
19 days and August in 2014 20 days), " model ice ice Li Zhiting " (correlation time point is August in 2014 19 days and August 20 in 2014
Day), " Chinese good sound Zhangjiang foam " (correlation time point be 2014 on August 19), " burning is retired " (correlation time point
For August in 2014 19 days) and " Meizu news conference " (correlation time point be 2014 on August 19) etc., and also searched for from user
Extracted in daily record foregoing each first association search word correlation time point and with " Garbo " co-occurrence number of times and respective visit capacity
(searching times for being scanned for using the association search word), wherein, " Heng Shi ground rice up-to-date event " is total to " Garbo "
Occurrence number is most, and respectively 13559 and 13429, visit capacity is respectively 26737 and 26168.
In step S103, according to the first association search word, correlation time point and its searching times that are extracted in step S102
And user search daily record chooses searching times and changes that amplitude exceedes the second association search word of preassigned, it is in search
Change direction and transformation period on number of times is interval, wherein, the change direction is positive or reverse.
Preferably, to any first association search word, following operation is performed：
The system of the log acquisition in the predetermined examination time period before its correlation time point is searched for according to the user
The searching times average of unit interval is counted, if the searching times of the first association search word are relative to the searching times
The change amplitude of average exceedes predetermined change threshold value, then the first association search selected ci poem is taken as into the second association search word, its
In, the change amplitude can be, but not limited to equal relative to the searching times with the searching times of the first association search word
The ratio of value represents that the predetermined change threshold value but can also be not limited to form of ratios, for example, it is assumed that the change width for calculating
It is 4 to spend, and predetermined change threshold value is 3, then its corresponding first association search selected ci poem is taken as into the second association search word.
Additionally, the searching times of the first association search word are made relative to the change direction of the searching times average
It is the change direction on searching times of the second association search word.If for example, the first association search word is searched
Rope number of times is more than the searching times average, then can determine change direction of the second association search word on searching times
It is positive direction, conversely, can then determine that change direction of the second association search word on searching times is negative direction.
It is its adjacent correlation time point is continuous or interruption is less than pre- timing to the second association search word of any selection
Between the correlation time point that is spaced merge that to turn into its transformation period interval.
Fig. 4 is the second association search word, its change direction on searching times for showing exemplary embodiment of the present
And the interval data instance figure of transformation period.
Assuming that the statistical unit time period is 1 day, it is 3 to make a reservation for change threshold value, and the examination time period is past 1 month.
Figure 4, it is seen that " Heng Shi ground rice up-to-date event " is equal in the searching times relative to the examination time period on the 19th of August in 2014
The change amplitude of value is 63.27, and searching times are that the searching times relative to the examination time period on the 20th of August in 26168,2014 are equal
The change amplitude of value is 20.45, and searching times are 26737, and " model ice ice Li Zhiting " is respectively 285.98 and 33.14, search time
Number is respectively 23983 and 30160, and " burning is retired " and " Meizu news conference " in August in 2014 19 days relative to examination
The change amplitude of the searching times average of time period is respectively 44.03 and 48.49, and searching times are respectively 30717 and 30405,
Thus the change direction of above-mentioned second association search word is forward direction.
In step S104, become according to the interval generation of the second association search word and transformation period chosen in step s 103
Gesture explains data.
It is provided in an embodiment of the present invention a kind of for generating the method that trend explains data, obtained by searching for daily record from user
The first association search word, correlation time point and its searching times are taken, then the second pass is filtered out from the first association search word
Connection search word, change direction and transformation period the interval generation trend according to the second association search word and its on searching times
Explain data, in this way, automatically generating objective and degree of accuracy trend higher explains data, provide the user with it is objective, effective, be
The trend of system is explained, and then strengthens Consumer's Experience.
Fig. 2 is that the user's generation trend for showing another exemplary embodiment of the present invention explains that the method flow of data is illustrated
Figure.The treatment of the step S101~step S103 in Fig. 2 is consistent with the treatment of corresponding steps in Fig. 1, will not be described in detail herein.
Reference picture 2, in step S105, the transformation period Interval Maps of the second association search word is waited to examine or check to described one group
On the search curve of search word.
Can be according within the examination time period, in one group of total search of search word to be examined or check described in each statistical unit time period
Number of times obtains the search curve.
In step S106, date back the search from the interval interval starting point of the transformation period of the second association search word and become
Change the stage start time point extended to along the change direction of the second association search word on curve, then calculate and closed described second
The transformation period for joining search word is interval interior described when the first average search number of times for examining or check search word and stage starting
Between the search word to be examined or check between point and interval initial time section the second average search number of times, wherein, association is searched
The number of times that rope word is searched within the statistical unit time period is used as its searching times.
It is change direction according to the second association search word on searching times, described first average in step S107
The value of searching times and the second average search number of times determines whether for the second association search selected ci poem to be taken as the 3rd association search
Word.
Exemplary embodiment of the invention, if the change direction is forward direction, and second average search
Number of times is less than the first average search number of times, then it is believed that the change direction of the second association search word is bent with the search
Line is matched in the change of the constant interval, and the second association search selected ci poem is taken as into the 3rd association search word；If described
Change direction is negative sense, and the second average search number of times is more than the first average search number of times, then by described second
Association search selected ci poem is taken as the 3rd association search word；If the change direction is positive and described second average search number of times
Not less than the first average search number of times, or if the change direction is negative sense and the second average search number of times
No more than described first average search number of times, then be not taken as the 3rd association search word by the second association search selected ci poem.
Also threshold value can be set by the value that the first average search number of times and the second average search number of times are differed
Further to control the selection of the 3rd association search word.In accordance with an alternative illustrative embodiment of the present invention, if the change side
To being forward direction, the second average search number of times is less than the first average search number of times, and first average search is secondary
Difference between several and the second average search number of times is more than the first predetermined searching times threshold value, then choose the association search
Word；If the change direction is negative sense, the second average search number of times is more than the first average search number of times, and institute
The difference stated between the second average search number of times and the first average search number of times is more than the second predetermined searching times threshold value, then
Choose the association search word；If the change direction is positive and the second average search number of times not less than described the
One average search number of times, then do not choose the association search word；If the change direction is forward direction, second average search
Number of times be less than the first average search number of times, and the first average search number of times and the second average search number of times it
Between difference be not more than the first predetermined searching times threshold value, then do not choose the association search word；If the change direction is
Negative sense and the second average search number of times is not more than the first average search number of times, then do not choose the association search
Word；If the change direction is negative sense, the second average search number of times is more than the first average search number of times, and institute
State the difference between the second average search number of times and the first average search number of times and be not more than the second predetermined searching times threshold value,
The association search word is not chosen then.
Fig. 5 is the trend matching schematic diagram of the second association search word for showing exemplary embodiment of the present.
As shown in figure 5, being example with the transformation period interval of the second association search word of the positive change on searching times
Illustrate, the point in black line frame as can be seen from the figure larger in figure represents the first average search number of times, less wire frame
In point represent the second average search number of times.As can be seen that the second average search number of times is less than first average search
Number of times, it can be considered that the second association search word is interval in this section of one group of search word to be examined or check in the transformation period
It is matching on search curve, you can be taken as the 3rd association search word with by the second association search selected ci poem.
Fig. 6 is the transformation period Interval Maps of the second association search word for showing exemplary embodiment of the present to waiting to examine or check
The data instance figure of the search curve of search word.
Reference picture 6, the partial data in addition to including the data shown in Fig. 4, also including interval matching judgment, from figure
In as can be seen that the transformation period of foregoing second association search word is interval and the one group of search word to be examined or check being mapped to search
Funicular curve is matched completely, the second association search selected ci poem can be taken as into the 3rd association search word.
In step S108, data are explained according to the 3rd association search word and transformation period interval generation trend.
Preferably, obtained from multiple data sources (such as news, variety show) according to the 3rd association search word and described
Data, and the searching times in its transformation period interval are obtained, according to the description data of the 3rd association search word
The interval trend of the transformation period is generated with searching times explain data.
Further, for interval the 3rd association search word for overlapping of transformation period, according to it in the transformation period area
Between be searched number of times from high to low order sequence.
Fig. 7 is the data trend of the generation for showing exemplary embodiment of the present and carries out trend with reference to other data sources
The exemplary plot of explanation, Fig. 8 is to show the related data sources exemplary plot corresponding with the data trend shown in Fig. 7.
Reference picture 7, search word to be examined or check is " young leaves Gu village ", and the examination time is 20140601~20140729, through sieving
The 3rd association search word extracted after choosing is " father where go the second season ", " the ancient village of young leaves " 5 days July in 2014 on the day of
Searching times are in peak in the examination time, herein the 3rd association search word " father where go the second season ", correlation time
On July 5th, 2014 and searching times 17336 can explain data as basic trend.Fig. 8 is shown according to other data sources
The description data of acquisition, explain that data are extended by the basic trend, and the volumes of searches in " young leaves Gu village " can exist
On July 5th, 2014, elevated trend was construed to suddenly, due to father go where the second season be with finding a view young leaves Gu village so that
The access trend in Xin Guye villages changes.
In sum, trend explains that data include but is not limited to association search word, time or time interval, association can be explained
The searching times of search word and the news of association search word or other related contents.
It is provided in an embodiment of the present invention a kind of for generating the method that trend explains data, obtained by searching for daily record from user
Association search word, correlation time point and its searching times are taken, and association search word is carried out by default standard or condition many
Secondary screening, trend explanation data are ultimately generated with reference to the searching times of correlation time point and its association search word, in this way, in this way,
Automatically generate objective and degree of accuracy trend higher and explain data, provide the user with objective, effective, system trend and explain,
And then strengthen Consumer's Experience.
Fig. 9 is the apparatus structure block diagram that data are explained for generating trend for showing an exemplary embodiment of the invention.Ginseng
According to Fig. 9, described device includes：Log acquisition unit 201, first information extraction unit 202, the second information extraction unit 203 with
And trend explains data generating unit 204.
Log acquisition unit 201 is used to obtain user's search daily record, wherein, user's search daily record is included from multiple
The internet hunt record of data source.
First information extraction unit 202 is used to be associated with one group of search word to be examined or check and searched from described extraction for searching for daily record
First association search word of rope, association search time point and its searching times.
According to a preferred embodiment of the invention, first information extraction unit 202 is used to be extracted from user search daily record
Any user is together searched for the search word any to be examined or check in one group of search word to be examined or check in a certain time interval
Search word, and/or in any query search word with one group of search word to be examined or check in search word any to be examined or check
The search word for together occurring as the first association search word,
According to a preferred embodiment of the invention, first information extraction unit 202 is additionally operable to according to predetermined timing statisticses section
Unit, the statistical unit time period that the first association search word is searched counts described as its correlation time point
The number of times that association search word is searched within the statistical unit time period is used as its searching times.
Second information extraction unit 203 is used for according to the first association search word, correlation time point and its search time for extracting
Several and user search daily record chooses searching times and changes that amplitude exceedes the second association search word of preassigned, it is searching
Change direction and transformation period on rope number of times is interval, and the change direction is for forward or backwards.
According to a preferred embodiment of the invention, the second information extraction unit 203 is used for any first association search
Word, performs following operation：According to user search log acquisition in the predetermined examination time period before its correlation time point
The statistical unit time period searching times average, if the searching times of the first association search word are relative to described
The change amplitude of searching times average exceedes predetermined change threshold value, then the first association search selected ci poem is taken as into the second association and searched
Rope word, and change direction using the searching times of the first association search word relative to the searching times average is used as institute
State change direction of the second association search word on searching times.
Preferably, the second information extraction unit 203 is additionally operable to the second association search word to any selection, and its is adjacent
Correlation time point is continuous or interval merges interval as its transformation period less than the correlation time point of predetermined time interval.
Trend explains that data generating unit 204 is used for according to the second association search word and the interval generation of transformation period
Trend explains data.
Preferably, described device also includes search word acquiring unit (not shown) to be examined or check, for from being examined or check
The unstructured data that the head stack or user that target word bag, user access are accessed extracts described one group and treats examination search
Word.
A kind of device that data are explained for generating trend provided in an embodiment of the present invention, is obtained by searching for daily record from user
The first association search word, correlation time point and its searching times are taken, then the second pass is filtered out from the first association search word
Connection search word, change direction and transformation period the interval generation trend solution according to the second association search word and its on searching times
Release data, so, it is possible to capture data trend Producing reason well, provide the user with one it is autonomous, effective, be
The trend of system is explained, and then strengthens Consumer's Experience.
Figure 10 is the apparatus structure block diagram that data are explained for generating trend for showing another exemplary embodiment of the present invention.
Reference picture 10, described device includes：Log acquisition unit 201, first information extraction unit 202, the second information extraction unit
203rd, the 3rd association search word acquiring unit 205 and trend explain data generating unit 204, wherein, log acquisition unit
201st, the information extraction unit 203 of first information extraction unit 202 and second is consistent with the corresponding units shown in Fig. 9, herein
No longer describe in detail.
3rd association search word acquiring unit 205 is used to arrive the transformation period Interval Maps of the second association search word
On the search change curve of one group of search word to be examined or check, the change of the searching times according to one group of search word to be examined or check
Filter out the 3rd association search word of change direction matching.
According to the preferred embodiment of the present invention, the 3rd association search word acquiring unit 205 is used to be closed for any described second
Connection search word, performs following operation：From the interval start time point backtracking that the transformation period of the second association search word is interval
The stage start time point extended to along the change direction of the second association search word on to the search change curve, calculates
First average search number of times of the search word to be examined or check and institute in the transformation period of the second association search word is interval
The second average search number of times of the search word to be examined or check between stage start time point and the interval start time point is stated, is used
Determine whether described according to the value of the change direction, the first average search number of times and the second average search number of times
Two association search selected ci poems are taken as the 3rd association search word.
If the change direction is forward direction, and the second average search number of times is less than first average search time
Number, then the 3rd association search word acquiring unit 205 chooses the association search word.
If the change direction is negative sense, and the second average search number of times is more than first average search
Number of times, then the 3rd association search word acquiring unit 205 choose the association search word.
If the change direction is not less than first average search for positive and described second average search number of times
Number of times, or if the change direction is for negative sense and the second average search number of times is not more than first average search
Number of times, then the 3rd association search word acquiring unit 205 do not choose the association search word.
Based on the 3rd association search word acquiring unit 205, now, trend explains that data generating unit 204 is used for according to institute
State the 3rd association search word and transformation period interval generation trend explains data.
Preferably, trend explains that data generating unit 204 is obtained according to the 3rd association search word from multiple data sources
Description data, and the searching times in its transformation period interval are obtained, and retouched according to the 3rd association search word
State data and searching times generate the interval trend of the transformation period and explain data.
Preferably, trend explains that data generating unit 204 is additionally operable to be searched for interval the 3rd association for overlapping of transformation period
Rope word, sorts according to its order in the interval searched number of times of the transformation period from high to low.
A kind of device that data are explained for generating trend provided in an embodiment of the present invention, is obtained by searching for daily record from user
Association search word, correlation time point and its searching times are taken, and association search word is carried out by default standard or condition many
Secondary screening, trend explanation data are ultimately generated with reference to the searching times of correlation time point and its association search word, be so, it is possible very
Data trend Producing reason is captured well, is provided the user with autonomous, effective, system a trend and is explained, and then
Enhancing Consumer's Experience.
It may be noted that the need for according to implementation, each step described in this application can be split as into more multi-step, also may be used
The part operation of two or more steps or step is combined into new step, to realize the purpose of the present invention.
Above-mentioned the method according to the invention can be realized in hardware, firmware, or be implemented as being storable in recording medium
Software or computer code in (such as CD ROM, RAM, floppy disk, hard disk or magneto-optic disk), or it is implemented through network download
Original storage in long-range recording medium or nonvolatile machine readable media and by the meter in being stored in local recording medium
Calculation machine code, so that method described here can be stored in uses all-purpose computer, application specific processor or programmable or special
With the such software processing in the recording medium of hardware (such as ASIC or FPGA).It is appreciated that computer, processor, micro-
Processor controller or programmable hardware include storing receive software or the storage assembly of computer code (for example, RAM,
ROM, flash memory etc.), when the software or computer code are by computer, processor or hardware access and execution, realize herein
The processing method of description.Additionally, when all-purpose computer accesses the code for the treatment for realizing being shown in which, the execution of code
All-purpose computer is converted into the special-purpose computer for performing the treatment being shown in which.
The above, specific embodiment only of the invention, but protection scope of the present invention is not limited thereto, and it is any
Those familiar with the art the invention discloses technical scope in, change or replacement can be readily occurred in, should all contain
Cover within protection scope of the present invention.Therefore, protection scope of the present invention should be based on the protection scope of the described claims.
Claims (20)
1. it is a kind of by the computer implemented method for generating trend explanation data, it is characterised in that methods described includes：
Obtain user's search daily record；
The first association search word, the correlation time with one group of search word association search to be examined or check are extracted from user search daily record
Point and its searching times；
Choose and search according to the first association search word, correlation time point and its searching times that extract and user search daily record
When rope number of times changes amplitude and exceedes the second association search word, its change direction on searching times and the change of preassigned
Between it is interval, the change direction is for forward or backwards；
Data are explained according to the second association search word and transformation period interval generation trend.
2. method according to claim 1, it is characterised in that methods described also includes：
The unstructured data that the head stack or user accessed from the target word bag, user to be examined or check are accessed extracts described one
Group search word to be examined or check.
3. method according to claim 2, it is characterised in that user's search daily record is included from multiple data sources
Internet hunt is recorded.
4. method according to claim 3, it is characterised in that methods described also includes：
The search of the transformation period Interval Maps of the second association search word to one group of search word to be examined or check is changed bent
On line, the change of the searching times according to one group of search word to be examined or check filters out the 3rd association search of change direction matching
Word, and
It is described to explain that the treatment of data includes according to the second association search word and transformation period interval generation trend：According to
The 3rd association search word and transformation period interval generation trend explain data.
5. method according to claim 4, it is characterised in that the search according to one group of search word to be examined or check time
The treatment that several changes filters out the 3rd association search word of change direction matching includes：
For any second association search word, following operation is performed：
The search change curve is dateed back from the interval interval start time point of the transformation period of the second association search word
On the stage start time point that is extended to along the change direction of the second association search word,
Calculate first average search time of the search word to be examined or check in the transformation period of the second association search word is interval
The second of the search word to be examined or check averagely is searched between several and the stage start time point and the interval start time point
Rope number of times,
Value according to the change direction, the first average search number of times and the second average search number of times determines whether will be described
Second association search selected ci poem is taken as the 3rd association search word.
6. method according to claim 5, it is characterised in that described averagely to be searched according to the change direction, described first
The value of rope number of times and the second average search number of times determines whether that the treatment for choosing the association search word includes：
If the change direction is forward direction, and the second average search number of times is less than the first average search number of times,
The association search word is then chosen,
If the change direction is negative sense, and the second average search number of times is more than the first average search number of times,
The association search word is then chosen,
If the change direction is not less than the first average search number of times for positive and described second average search number of times,
Or if the change direction is for negative sense and the second average search number of times is not more than the first average search number of times,
The association search word is not chosen then.
7. method according to claim 5, it is characterised in that described averagely to be searched according to the change direction, described first
The value of rope number of times and the second average search number of times determines whether that the treatment for choosing the association search word includes：
If the change direction is forward direction, the second average search number of times is less than the first average search number of times, and
Difference between the first average search number of times and the second average search number of times is more than the first predetermined searching times threshold value,
The association search word is then chosen,
If the change direction is negative sense, the second average search number of times is more than the first average search number of times, and
Difference between the second average search number of times and the first average search number of times is more than the second predetermined searching times threshold value,
The association search word is then chosen,
If the change direction is not less than the first average search number of times for positive and described second average search number of times,
Do not choose the association search word then,
If the change direction is forward direction, the second average search number of times is less than the first average search number of times, and
Difference between the first average search number of times and the second average search number of times is not more than the first predetermined searching times threshold
Value, then do not choose the association search word,
If the change direction is for negative sense and the second average search number of times is not more than the first average search number of times,
Do not choose the association search word then,
If the change direction is negative sense, the second average search number of times is more than the first average search number of times, and
Difference between the second average search number of times and the first average search number of times is not more than the second predetermined searching times threshold
Value, then do not choose the association search word.
8. the method according to any one of claim 1~7, it is characterised in that described to be carried from user search daily record
Take the treatment bag with the first association search word, correlation time point and its searching times of one group of search word association search to be examined or check
Include：
From the user search daily record extract any user in a certain time interval with one group of search word to be examined or check in
The search word that any search word to be examined or check together is searched for, and/or wait to examine or check with described one group in any query search word
The search word that search word any to be examined or check in search word together occurs as the first association search word,
According to predetermined timing statisticses section unit, the statistical unit time period that the first association search word is searched as its
Correlation time point, and the searched number of times within the statistical unit time period of the association search word is counted as its search
Number of times.
9. method according to claim 8, it is characterised in that described extraction from user search daily record is needed checking with a group
The treatment for looking into the first association search word, correlation time point and its searching times of search word association search also includes：
Such search word is deleted from the first association search word for extracting：Its with one group of search word to be examined or check in any treat
The number of times that examination search word together occurs is less than predetermined co-occurrence number of times.
10. method according to claim 9, it is characterised in that it is described according to the first association search word for extracting, association when
Between point and its searching times and the user search daily record choose searching times change amplitude exceed preassigned second pass
The interval treatment of connection search word, its change direction and transformation period on searching times includes：
To any first association search word, following operation is performed：
The statistics list of the log acquisition in the predetermined examination time period before its correlation time point is searched for according to the user
The searching times average of position time period,
If the searching times of the first association search word exceed predetermined relative to the change amplitude of the searching times average
Change threshold value, then the first association search selected ci poem is taken as the second association search word, and by the first association search word
Change direction of the searching times relative to the searching times average as the second association search word on searching times
Change direction,
It is its adjacent correlation time point is continuous or interval is less than predetermined time interval to the second association search word of any selection
Correlation time point merge that to turn into its transformation period interval.
11. method according to any one of claim 4~7, it is characterised in that described according to the 3rd association search
Word and transformation period interval generation trend explain that the treatment of data includes：
Description data are obtained from multiple data sources according to the 3rd association search word, and is obtained in its transformation period interval
Searching times,
The description data and searching times according to the 3rd association search word generate the interval trend of the transformation period
Explain data.
12. methods according to claim 11, it is characterised in that described according to the 3rd association search word and change
Time interval generation trend explains that the treatment of data also includes：
For interval the 3rd association search word for overlapping of transformation period, according to it in the interval searched number of times of the transformation period
Order sequence from high to low.
13. a kind of devices that data are explained for generating trend, it is characterised in that described device includes：
Log acquisition unit, for obtaining user's search daily record；
First information extraction unit, for extracting the first pass with one group of search word association search to be examined or check from the search daily record
Connection search word, association search time point and its searching times；
Second information extraction unit, for the first association search word, the correlation time point that are extracted according to first information extraction unit
And its searching times and user search daily record are chosen searching times change amplitude and are searched more than the second association of preassigned
Rope word, its change direction and transformation period on searching times are interval, and the change direction is for forward or backwards；
Trend explains data generating unit, for according to the second association search word and transformation period interval generation trend solution
Release data.
14. devices according to claim 13, it is characterised in that described device also includes：
Search word acquiring unit to be examined or check, head stack or user for being accessed from the target word bag, user to be examined or check are visited
The unstructured data asked extracts one group of search word to be examined or check.
15. devices according to claim 14, it is characterised in that described device also includes：
3rd association search word acquiring unit, for by the transformation period Interval Maps of the second association search word to described
On the search change curve of group search word to be examined or check, the change of the searching times according to one group of search word to be examined or check is filtered out
3rd association search word of change direction matching, and
The trend explains that data generating unit is used to become according to the 3rd association search word and the interval generation of transformation period
Gesture explains data.
16. devices according to claim 15, it is characterised in that the 3rd association search word acquiring unit be used for for
Any second association search word, performs following operation：
The search change curve is dateed back from the interval interval start time point of the transformation period of the second association search word
On the stage start time point that is extended to along the change direction of the second association search word,
Calculate first average search time of the search word to be examined or check in the transformation period of the second association search word is interval
The second of the search word to be examined or check averagely is searched between several and the stage start time point and the interval start time point
Rope number of times,
Value according to the change direction, the first average search number of times and the second average search number of times determines whether will be described
Second association search selected ci poem is taken as the 3rd association search word.
17. devices according to claim 16, it is characterised in that
If the change direction is forward direction, and the second average search number of times is less than the first average search number of times,
Then the 3rd association search word acquiring unit chooses the association search word,
If the change direction is negative sense, and the second average search number of times is more than the first average search number of times,
Then the 3rd association search word acquiring unit chooses the association search word,
If the change direction is not less than the first average search number of times for positive and described second average search number of times,
Or if the change direction is for negative sense and the second average search number of times is not more than the first average search number of times,
Then the 3rd association search word acquiring unit does not choose the association search word.
18. device according to any one of claim 13~17, it is characterised in that the first information extraction unit is used
In：
From the user search daily record extract any user in a certain time interval with one group of search word to be examined or check in
The search word that any search word to be examined or check together is searched for, and/or wait to examine or check with described one group in any query search word
The search word that search word any to be examined or check in search word together occurs as the first association search word,
According to predetermined timing statisticses section unit, the statistical unit time period that the first association search word is searched as its
Correlation time point, and the searched number of times within the statistical unit time period of the association search word is counted as its search
Number of times.
19. devices according to claim 18, it is characterised in that second information extraction unit is used for：
To any first association search word, following operation is performed：
The statistics list of the log acquisition in the predetermined examination time period before its correlation time point is searched for according to the user
The searching times average of position time period,
If the searching times of the first association search word exceed predetermined relative to the change amplitude of the searching times average
Change threshold value, then the first association search selected ci poem is taken as the second association search word, and by the first association search word
Change direction of the searching times relative to the searching times average as the second association search word on searching times
Change direction,
Second information extraction unit is additionally operable to the second association search word to any selection, and its adjacent correlation time point is continuous
Or interval merges interval as its transformation period less than the correlation time point of predetermined time interval.
20. device according to any one of claim 15~17, it is characterised in that the trend explains data genaration list
Unit is used for：
Description data are obtained from multiple data sources according to the 3rd association search word, and is obtained in its transformation period interval
Searching times,
The description data and searching times according to the 3rd association search word generate the interval trend of the transformation period
Explain data,
Wherein, for interval the 3rd association search word for overlapping of transformation period, the trend explains that data generating unit is additionally operable to
Sorted according to its order in the interval searched number of times of the transformation period from high to low.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410652571.3A CN104331493B (en) | 2014-11-17 | 2014-11-17 | By the computer implemented method and device that data are explained for generating trend |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410652571.3A CN104331493B (en) | 2014-11-17 | 2014-11-17 | By the computer implemented method and device that data are explained for generating trend |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104331493A CN104331493A (en) | 2015-02-04 |
| CN104331493B true CN104331493B (en) | 2017-07-07 |
Family
ID=52406220
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410652571.3A Active CN104331493B (en) | 2014-11-17 | 2014-11-17 | By the computer implemented method and device that data are explained for generating trend |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104331493B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105302867B (en) * | 2015-09-28 | 2019-06-11 | 浙江宇视科技有限公司 | A kind of search engine inquiry method and device |
| US10157240B2 (en) * | 2015-10-01 | 2018-12-18 | Ebay Inc. | Systems and methods to generate a concept graph |
| CN107256244B (en) * | 2017-06-01 | 2021-09-03 | 北京京东尚科信息技术有限公司 | Data processing method and system |
| CN107273508B (en) * | 2017-06-20 | 2020-07-10 | 北京百度网讯科技有限公司 | Information processing method and device based on artificial intelligence |
| CN110019367B (en) * | 2017-12-28 | 2022-04-12 | 北京京东尚科信息技术有限公司 | Method and device for counting data characteristics |
| CN110688846B (en) * | 2018-07-06 | 2023-11-07 | 北京京东尚科信息技术有限公司 | Periodic word mining method, system, electronic equipment and readable storage medium |
| CN110334277B (en) * | 2019-06-28 | 2020-08-21 | 北京天眼查科技有限公司 | User search behavior identification method and device |
| CN113535990B (en) * | 2020-11-10 | 2023-12-15 | 腾讯科技（深圳）有限公司 | Method, device, storage medium and electronic equipment for determining multimedia content |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011090036A1 (en) * | 2010-01-19 | 2011-07-28 | 日本電気株式会社 | Trend information retrieval device, trend information retrieval method and recording medium |
| CN103324718A (en) * | 2013-06-25 | 2013-09-25 | 百度在线网络技术（北京）有限公司 | Topic venation digging method and system based on massive searching logs |
| CN103500163A (en) * | 2013-07-24 | 2014-01-08 | 百度在线网络技术（北京）有限公司 | Method and device for recognizing event key progress |
-
2014
- 2014-11-17 CN CN201410652571.3A patent/CN104331493B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011090036A1 (en) * | 2010-01-19 | 2011-07-28 | 日本電気株式会社 | Trend information retrieval device, trend information retrieval method and recording medium |
| CN103324718A (en) * | 2013-06-25 | 2013-09-25 | 百度在线网络技术（北京）有限公司 | Topic venation digging method and system based on massive searching logs |
| CN103500163A (en) * | 2013-07-24 | 2014-01-08 | 百度在线网络技术（北京）有限公司 | Method and device for recognizing event key progress |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104331493A (en) | 2015-02-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104331493B (en) | By the computer implemented method and device that data are explained for generating trend | |
| CN108241667B (en) | Method and apparatus for pushed information | |
| KR101536520B1 (en) | Method and server for extracting topic and evaluating compatibility of the extracted topic | |
| US10032081B2 (en) | Content-based video representation | |
| CN109033200B (en) | Event extraction method, device, equipment and computer readable medium | |
| CN103577593B (en) | A kind of video aggregation method and system based on microblog hot topic | |
| Bhattacharya et al. | Deep twitter diving: Exploring topical groups in microblogs at scale | |
| US20150149383A1 (en) | Method and device for acquiring product information, and computer storage medium | |
| US20150205580A1 (en) | Method and System for Sorting Online Videos of a Search | |
| WO2013059290A1 (en) | Sentiment and influence analysis of twitter tweets | |
| CN106599215A (en) | Question generation method and question generation system based on deep learning | |
| US11036818B2 (en) | Method and system for detecting graph based event in social networks | |
| Pota et al. | A subword-based deep learning approach for sentiment analysis of political tweets | |
| CN104881495B (en) | Folder path identification and folder cleaning method and device | |
| CN105512300B (en) | information filtering method and system | |
| CN105677802A (en) | Internet information analysis system | |
| JP2008203933A (en) | Category creation method and apparatus and document classification method and apparatus | |
| CN105843889A (en) | Credibility based big data and general data oriented data collection method and system | |
| Zarrad et al. | The evaluation of the public opinion-a case study: Mers-cov infection virus in ksa | |
| CN104182539A (en) | Abnormal information batch processing method and system | |
| KR101780237B1 (en) | Method and device for answering user question based on q&a data provided on online | |
| KR101727686B1 (en) | Method for extracting semantic entity topic | |
| Cuzzola et al. | Filtering inaccurate entity co-references on the linked open data | |
| Khan et al. | The presence of Twitter bots and cyborgs in the# FeesMustFall campaign | |
| CN105843890A (en) | Knowledge base based big data and general data oriented data collection method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |