CN109615013A - The Sensor Events flow point segmentation method of Behavior-based control characteristic probability distribution constraint - Google Patents
The Sensor Events flow point segmentation method of Behavior-based control characteristic probability distribution constraint Download PDFInfo
- Publication number
- CN109615013A CN109615013A CN201811535743.3A CN201811535743A CN109615013A CN 109615013 A CN109615013 A CN 109615013A CN 201811535743 A CN201811535743 A CN 201811535743A CN 109615013 A CN109615013 A CN 109615013A
- Authority
- CN
- China
- Prior art keywords
- behavior
- probability distribution
- count
- probability
- sensor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides a kind of Sensor Events flow point segmentation method of Behavior-based control characteristic probability distribution constraint, comprising: calculates behavior probability feature；As unit of day, the transfering probability distribution of any two behavior in training set is calculated；Calculate the probability distribution of the daily frequency of occurrence of each behavior；Calculate the probability distribution of the number of each behavior trigger sensor；Calculate the probability distribution of the number of species of each behavior trigger sensor；Calculate the probability distribution of the behavior instance number of each cluster；Calculate the probability distribution of the behavior instance number of each cluster；The probability characteristics collection and segmentation Sensor Events stream of constituting action.Resident's daily behavior recognition methods of Behavior-based control characteristic probability distribution constraint proposed by the present invention by calculating the probability distribution of a variety of behavioural characteristics, and the method for applied probability constraint solving can in the hope of meet feature set probability distribution optimal sensor event behavior example segmentation result.
Description
Technical field
The present invention relates to Sensor Events segmentation technologies, specifically, more particularly to Behavior-based control characteristic probability point
The Sensor Events flow point segmentation method of cloth constraint.
Background technique
Resident's daily behavior will continuously trigger one group of sensor, ideally, the Sensor Events being continuously triggered
Stream should be split according to complete behavior example.But the main method of moment sensor flow of event segmentation is specified first
Then one time window length or spatial window (number of sensors) length are averaged Ground Split sensor according to length of window
Flow of event.The primitive character of daily behavior identification is Sensor Events stream, and the Sensor Events stream based on fixed length of window is drawn
Point method, will lead to that same behavior example is divided in multiple and different windows or multiple behavior examples are divided in together
One window, such division result perhaps cause the feature of daily behavior imperfect or are done by other behavioural characteristics
It disturbs, the effect for causing daily behavior to identify is poor.
Summary of the invention
According to technical problem set forth above, and provide the Sensor Events flow point of Behavior-based control characteristic probability distribution constraint
Segmentation method.
The present invention is based on the Sensor Events flow point segmentation methods of behavioural characteristic probability distribution constraint, which is characterized in that at least
Include: S1: calculating behavior probability feature；S2: segmentation Sensor Events stream.
Further, the step S1 behavior probability feature calculation
S11: as unit of day, the transfering probability distribution of any two behavior in training set is calculated；
S12: the probability distribution of the daily frequency of occurrence of each behavior is calculated；
S13: the probability distribution of the number of each behavior trigger sensor is calculated；
S14: the probability distribution of the number of species of each behavior trigger sensor is calculated；
S15: being unified for numerical value in seconds for the duration of each behavior, then uses clustering algorithm pair
The duration of behavior clusters, and calculates the probability distribution of the behavior instance number of each cluster；
S16: will be unified for numerical value in seconds at the beginning of each behavior, then using clustering algorithm to behavior
At the beginning of cluster, calculate the probability distribution of the behavior instance number of each cluster；
S17: pass through the probability characteristics collection of the feature constituting action of step S11- step S16.
Further, the Sensor Events stream segmentation at least also comprises the steps of:
S21: initialization maximum probability distribution variable C is 0；
S22: counting daily behavior example border sensor in training set, i.e. first sensor being triggered and last
One sensor being triggered；
S23: according to the border sensor, it is real to obtain k initial behavior for the cutting Sensor Events stream in test set
Example boundary；
S24: the probability distribution of quantity occurs for daily behavior in the trained centralized calculation one day；
Assuming that date set is denoted as D in training set, count (D) indicates the quantity on date in D；Then for d ∈ D, count
(d) indicate that the behavior instance number occurred in date d, the type set of count (d) are denoted as N；For n ∈ N, it is real that n times behavior occurs
The number of days of example is denoted as count (n)；
Then in one day daily behavior occur quantity probability are as follows:
P (n, D)=count (n)/count (D)；
S25: the behavior example of random selection quantity n；
S26: the probability characteristics collection obtained according to the step S17, by being based on probability distribution constraint solving strategy, in institute
Selection in K initial boundary of test set is stated so that constraining n boundary of maximization；
S27: if maximum constrained probability is greater than C, the maximum constrained probability replaces the value of C*, and from the daily row
Deleted in probability distribution for quantity occurs current behavior instance number and probability, and go to step S25；If maximum constrained is general
Rate is less than or equal to C, then whether the probability distribution that quantity occurs for daily behavior is sky, then terminates.
Further, the transfering probability distribution for calculating any two behavior in training set:
If behavior category set is A, for behavior any two behavior a, b ∈ A, wherein<a, b>expression b behavior example
For the adjacent subsequent of a behavior example, count (<a, x>) indicate a in training set it is adjacent it is subsequent be x number；Then arbitrary act
The probability that a is shifted to b are as follows:
P (a, b)=count (<a, b>)/|<a, x>| x ∈ A } | (1).
Further, the probability distribution for calculating each daily frequency of occurrence of behavior:
Assuming that behavior category set is combined into D for the collection on date in A and training set, then for date d ∈ D, for behavior a ∈
A and designated date d ∈ D, probability of the behavior a in date d frequency of occurrence are as follows:
P (a, d)=count (a, d)/count (A, d) (2)；
Wherein count (A, d) indicates that the total degree of behavior occurs in date d；Count (a, d) indicates that a goes out occurrence in date d
Number.
Further, the probability distribution of the number for calculating each behavior trigger sensor:
Assuming that behavior category set is A, for behavior a ∈ A, the instance number that a occurs in training set is count (a)；a
Example collection be denoted as AIa, for aia∈AIa, for n ∈ Na, the number of behavior a trigger sensor is that the instance number of n is denoted as
Count (a, n), then the number of a trigger sensor is the probability of n are as follows:
P (a, n)=count (a, n)/count (a) (3)；
Wherein, sensors (aia) indicate aiaThe quantity of trigger sensor, sensors (aia) type set be denoted as Na。
Further, the clustering algorithm clusters to the duration of behavior, calculates the behavior example of each cluster
Several probability distribution:
Assuming that behavior category set is A, the example collection of A is denoted as AI, indicates that ai is held for ai ∈ AI, duration (ai)
The continuous time；
By K mean cluster algorithm to set duration (ai) | ai ∈ AIa, cluster result is denoted asCount (c) is the behavior instance number for including in cluster c, and count (AI) is the behavior instance number in AI；Each
The probability of the behavior instance number of cluster c ∈ C are as follows:
P (c)=count (c)/count (AI) (5).
The present invention has the advantages that resident's daily behavior of Behavior-based control characteristic probability distribution constraint proposed by the present invention is known
Other method is by calculating the probability distribution of a variety of behavioural characteristics, and the method for applied probability constraint solving can be in the hope of meeting feature
Collect the behavior example segmentation result of the optimal sensor event of probability distribution.Compared to the Sensor Events of more existing fixed window
Flow point segmentation method, can be closer to the segmentation result of true behavior example, to be subsequent promotion behavior using the present invention
The effect of identification, which provides, preferably to be supported.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to do simply to introduce, it should be apparent that, the accompanying drawings in the following description is this hair
Bright some embodiments for those of ordinary skill in the art without any creative labor, can be with
It obtains other drawings based on these drawings.
Fig. 1 is behavior probability feature calculation overall flow schematic diagram of the present invention.
Fig. 2 is that inventive sensor flow of event divides overall flow schematic diagram.
Specific embodiment
In order to enable those skilled in the art to better understand the solution of the present invention, below in conjunction in the embodiment of the present invention
Attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only
The embodiment of a part of the invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill people
The model that the present invention protects all should belong in member's every other embodiment obtained without making creative work
It encloses.
It should be noted that description and claims of this specification and term " first " in above-mentioned attached drawing, "
Two " etc. be to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should be understood that using in this way
Data be interchangeable under appropriate circumstances, so as to the embodiment of the present invention described herein can in addition to illustrating herein or
Sequence other than those of description is implemented.In addition, term " includes " and " having " and their any deformation, it is intended that cover
Cover it is non-exclusive include, for example, the process, method, system, product or equipment for containing a series of steps or units are not necessarily limited to
Step or unit those of is clearly listed, but may include be not clearly listed or for these process, methods, product
Or other step or units that equipment is intrinsic.
As shown in Figs. 1-2, the present invention includes that a kind of Sensor Events stream of Behavior-based control characteristic probability distribution constraint is divided
Method includes at least: S1: calculating behavior probability feature；S2: segmentation Sensor Events stream.
As preferred embodiment, step S1 behavior probability feature calculation
S11: as unit of day, the transfering probability distribution of any two behavior in training set is calculated；
In the present embodiment, if behavior category set is A, for behavior any two behavior a, b ∈ A, wherein < a, b
>indicate b behavior example be a behavior example it is adjacent subsequent, count (<a, x>) indicate training set in a it is adjacent it is subsequent be x
Number；The then probability that arbitrary act a is shifted to b are as follows:
P (a, b)=count (<a, b>)/|<a, x>| x ∈ A } | (1).
S12: the probability distribution of the daily frequency of occurrence of each behavior is calculated；
In the present embodiment, it is assumed that behavior category set is that the collection on date in A and training set is combined into D, then for date d
∈ D, for behavior a ∈ A and designated date d ∈ D, probability of the behavior a in date d frequency of occurrence are as follows:
P (a, d)=count (a, d)/count (A, d) (2)；
Wherein count (A, d) indicates that the total degree of behavior occurs in date d；Count (a, d) indicates that a goes out occurrence in date d
Number.
S13: the probability distribution of the number of each behavior trigger sensor is calculated；
As preferred embodiment, it is assumed that behavior category set is A, for behavior a ∈ A, what a occurred in training set
Instance number is count (a)；The example collection of a is denoted as AIa, for aia∈AIa, for n ∈ Na, of behavior a trigger sensor
Number is that the instance number of n is denoted as count (a, n), then the number of a trigger sensor is the probability of n are as follows:
P (a, n)=count (a, n)/count (a) (3)；
Wherein, sensors (aia) indicate aiaThe quantity of trigger sensor, sensors (aia) type set be denoted as Na。
S14: the probability distribution of the number of species of each behavior trigger sensor is calculated；
S15: being unified for numerical value in seconds for the duration of each behavior, then using clustering algorithm to behavior
Duration cluster, calculate the probability distribution of the behavior instance number of each cluster；
As preferred embodiment, it is assumed that behavior category set is A, and the example collection of A is denoted as AI, for ai ∈ AI,
Duration (ai) indicates ai duration；
By K mean cluster algorithm to set duration (ai) | ai ∈ AIa, cluster result is denoted asCount (c) is the behavior instance number for including in cluster c, and count (AI) is the behavior instance number in AI；Each
The probability of the behavior instance number of cluster c ∈ C are as follows:
P (c)=count (c)/count (AI) (5).
S16: will be unified for numerical value in seconds at the beginning of each behavior, then using clustering algorithm to behavior
At the beginning of cluster, calculate the probability distribution of the behavior instance number of each cluster；
S17: pass through the probability characteristics collection of the feature constituting action of step S11- step S16.
As preferred embodiment, the segmentation of Sensor Events stream is at least also comprised the steps of:
S21: initialization maximum probability distribution variable C is 0；
S22: counting daily behavior example border sensor in training set, i.e. first sensor being triggered and last
One sensor being triggered；
S23: according to border sensor, the cutting Sensor Events stream in test set obtains k initial behavior example side
Boundary；
S24: the probability distribution of quantity occurs for daily behavior in training centralized calculation one day；
Assuming that date set is denoted as D in training set, count (D) indicates the quantity on date in D；Then for d ∈ D, count
(d) indicate that the behavior instance number occurred in date d, the type set of count (d) are denoted as N；For n ∈ N, it is real that n times behavior occurs
The number of days of example is denoted as count (n)；
Then in one day daily behavior occur quantity probability are as follows:
P (n, D)=count (n)/count (D)；
S25: the behavior example of random selection quantity n；
S26: the probability characteristics collection obtained according to step S17, by being based on probability distribution constraint solving strategy, in test set
Selection in K initial boundary is so that constrain n boundary of maximization；
S27: if maximum constrained probability is greater than C, maximum constrained probability replaces the value of C*, and number occurs from daily behavior
Deleted in the probability distribution of amount current behavior instance number and probability, and go to step S25；If maximum constrained probability be less than etc.
In C, then whether the probability distribution that quantity occurs for daily behavior is sky, then terminates.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
In the above embodiment of the invention, it all emphasizes particularly on different fields to the description of each embodiment, does not have in some embodiment
The part of detailed description, reference can be made to the related descriptions of other embodiments.
In several embodiments provided herein, it should be understood that disclosed technology contents can pass through others
Mode is realized.Wherein, the apparatus embodiments described above are merely exemplary, such as the division of the unit, Ke Yiwei
A kind of logical function partition, there may be another division manner in actual implementation, for example, multiple units or components can combine or
Person is desirably integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual
Between coupling, direct-coupling or communication connection can be through some interfaces, the INDIRECT COUPLING or communication link of unit or module
It connects, can be electrical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
On unit.It can some or all of the units may be selected to achieve the purpose of the solution of this embodiment according to the actual needs.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
Equipment (can for personal computer, server or network equipment etc.) execute each embodiment the method for the present invention whole or
Part steps.And storage medium above-mentioned includes: that USB flash disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
Reservoir (RAM, Random Access Memory), mobile hard disk, magnetic or disk etc. be various to can store program code
Medium.
Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations；To the greatest extent
Pipe present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement；And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme.
Claims (6)
1. the Sensor Events flow point segmentation method of Behavior-based control characteristic probability distribution constraint, which is characterized in that include at least: S1:
Calculate behavior probability feature；S2: segmentation Sensor Events stream；
The step S1 behavior probability feature calculation
S11: as unit of day, the transfering probability distribution of any two behavior in training set is calculated；
S12: the probability distribution of the daily frequency of occurrence of each behavior is calculated；
S13: the probability distribution of the number of each behavior trigger sensor is calculated；
S14: the probability distribution of the number of species of each behavior trigger sensor is calculated；
S15: being unified for numerical value in seconds for the duration of each behavior, then using clustering algorithm to behavior
Duration cluster, calculate the probability distribution of the behavior instance number of each cluster；
S16: it will be unified for numerical value in seconds at the beginning of each behavior, then behavior is opened using clustering algorithm
Moment beginning clusters, and calculates the probability distribution of the behavior instance number of each cluster；
S17: pass through the probability characteristics collection of the feature constituting action of step S11- step S16.
2. the Sensor Events flow point segmentation method of Behavior-based control characteristic probability distribution constraint according to claim 1, special
Sign also resides in:
The Sensor Events stream segmentation at least also comprises the steps of:
S21: initialization maximum probability distribution variable C is 0；
S22: in training set count daily behavior example border sensor, i.e. first sensor being triggered and the last one
The sensor being triggered；
S23: according to the border sensor, the cutting Sensor Events stream in test set obtains k initial behavior example side
Boundary；
S24: the probability distribution of quantity occurs for daily behavior in the trained centralized calculation one day；
Assuming that date set is denoted as D in training set, count (D) indicates the quantity on date in D；Then for d ∈ D, count (d) table
Show that, in the behavior instance number that date d occurs, the type set of count (d) is denoted as N；For n ∈ N, n times behavior example occurs
Number of days is denoted as count (n)；
Then in one day daily behavior occur quantity probability are as follows:
P (n, D)=count (n)/count (D)；
S25: the behavior example of random selection quantity n；
S26: the probability characteristics collection obtained according to the step S17, by being based on probability distribution constraint solving strategy, in the survey
The n boundary so that constraint maximization is selected in K initial boundary of examination collection；
S27: if maximum constrained probability is greater than C, the maximum constrained probability replaces the value of C*, and sends out from the daily behavior
Deleted in the probability distribution of raw quantity current behavior instance number and probability, and go to step S25；If maximum constrained probability is small
In being equal to C, then whether the probability distribution that quantity occurs for daily behavior is sky, then terminates.
3. the Sensor Events flow point segmentation method of Behavior-based control characteristic probability distribution constraint according to claim 1, special
Sign also resides in:
The transfering probability distribution of any two behavior in the calculating training set:
If behavior category set is A, for behavior any two behavior a, b ∈ A, wherein<a, b>expression b behavior example are a row
For the adjacent subsequent of example, count (<a, x>) indicate a in training set it is adjacent it is subsequent be x number；Then arbitrary act a is to b
The probability of transfer are as follows:
P (a, b)=count (<a, b>)/|<a, x>| x ∈ A } | (1).
4. the Sensor Events flow point segmentation method of Behavior-based control characteristic probability distribution constraint according to claim 1, special
Sign also resides in:
The probability distribution for calculating each daily frequency of occurrence of behavior:
Assuming that behavior category set is that the collection on date in A and training set is combined into D, then for date d ∈ D, for behavior a ∈ A and
The probability of designated date d ∈ D, behavior a in date d frequency of occurrence are as follows:
P (a, d)=count (a, d)/count (A, d) (2)；
Wherein count (A, d) indicates that the total degree of behavior occurs in date d；Count (a, d) indicates a in date d frequency of occurrence.
5. the Sensor Events flow point segmentation method of Behavior-based control characteristic probability distribution constraint according to claim 1, special
Sign also resides in:
The probability distribution of the number for calculating each behavior trigger sensor:
Assuming that behavior category set is A, for behavior a ∈ A, the instance number that a occurs in training set is count (a)；The reality of a
Example set is denoted as AIa, for aia∈AIa, for n ∈ Na, the number of behavior a trigger sensor is that the instance number of n is denoted as count
(a, n), then the number of a trigger sensor is the probability of n are as follows:
P (a, n)=count (a, n)/count (a) (3)；
Wherein, sensors (aia) indicate aiaThe quantity of trigger sensor, sensors (aia) type set be denoted as Na。
6. the Sensor Events flow point segmentation method of Behavior-based control characteristic probability distribution constraint according to claim 1, special
Sign also resides in:
The clustering algorithm clusters to the duration of behavior, calculates the probability distribution of the behavior instance number of each cluster:
Assuming that behavior category set is A, the example collection of A is denoted as AI, indicates that ai is lasting for ai ∈ AI, duration (ai)
Time；
By K mean cluster algorithm to set duration (ai) | ai ∈ AIa, cluster result is denoted as
Count (c) is the behavior instance number for including in cluster c, and count (AI) is the behavior instance number in AI；The behavior of each cluster c ∈ C
The probability of instance number are as follows:
P (c)=count (c)/count (AI) (5).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811535743.3A CN109615013B (en) | 2018-12-14 | 2018-12-14 | Sensor event stream segmentation method based on behavior characteristic probability distribution constraint |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811535743.3A CN109615013B (en) | 2018-12-14 | 2018-12-14 | Sensor event stream segmentation method based on behavior characteristic probability distribution constraint |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109615013A true CN109615013A (en) | 2019-04-12 |
| CN109615013B CN109615013B (en) | 2022-09-27 |
Family
ID=66010051
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811535743.3A Active CN109615013B (en) | 2018-12-14 | 2018-12-14 | Sensor event stream segmentation method based on behavior characteristic probability distribution constraint |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109615013B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130185232A1 (en) * | 2012-01-17 | 2013-07-18 | International Business Machines Corporation | Probabilistic event networks based on distributed time-stamped data |
| CN107392311A (en) * | 2016-05-17 | 2017-11-24 | 阿里巴巴集团控股有限公司 | The method and apparatus of sequence cutting |
-
2018
- 2018-12-14 CN CN201811535743.3A patent/CN109615013B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130185232A1 (en) * | 2012-01-17 | 2013-07-18 | International Business Machines Corporation | Probabilistic event networks based on distributed time-stamped data |
| CN107392311A (en) * | 2016-05-17 | 2017-11-24 | 阿里巴巴集团控股有限公司 | The method and apparatus of sequence cutting |
Non-Patent Citations (1)
| Title |
|---|
| 吴拨荣等: "置信传播和模拟退火相结合求解约束满足问题", 《计算机应用研究》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109615013B (en) | 2022-09-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108427708B (en) | Data processing method, data processing apparatus, storage medium, and electronic apparatus | |
| Kim et al. | Improved churn prediction in telecommunication industry by analyzing a large network | |
| CN105740268B (en) | A kind of information-pushing method and device | |
| CN108090433A (en) | Face identification method and device, storage medium, processor | |
| CN109173263A (en) | A kind of image processing method and device | |
| CN108989888A (en) | Video content playback method, device, smart machine and storage medium | |
| CN107239993A (en) | A kind of matrix decomposition recommendation method and system based on expansion label | |
| CN105260414B (en) | User behavior similarity calculation method and device | |
| CN108228844A (en) | A kind of picture screening technique and device, storage medium, computer equipment | |
| CN110113634A (en) | A kind of information interaction method, device, equipment and storage medium | |
| CN106202304A (en) | Method of Commodity Recommendation based on video and device | |
| CN109785066A (en) | Products Show method, apparatus, equipment and storage medium based on micro- expression | |
| CN109597858A (en) | A kind of classification method of trade company and its recommended method and its device of device and trade company | |
| CN110119707A (en) | A kind of human motion recognition method | |
| CN109636482A (en) | Data processing method and system based on similarity model | |
| CN108197795A (en) | The account recognition methods of malice group, device, terminal and storage medium | |
| CN115049076A (en) | Iterative clustering type federal learning method based on prototype network | |
| CN103336800B (en) | The method of the fingerprint storage comparison that Behavior-based control is analyzed | |
| CN106169961A (en) | The network parameter processing method and processing device of neutral net based on artificial intelligence | |
| CN106682058A (en) | Screening method, device and system of application programs | |
| León et al. | Artificial Intelligence in Agriculture | |
| CN109993586A (en) | A kind of method and apparatus for product of marketing | |
| CN109146606A (en) | A kind of brand recommended method, electronic equipment, storage medium and system | |
| CN109615013A (en) | The Sensor Events flow point segmentation method of Behavior-based control characteristic probability distribution constraint | |
| CN106933904B (en) | The filter method and device of data |
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 |