CN106251070A - A kind of method and device of automatic acquisition early warning object - Google Patents
A kind of method and device of automatic acquisition early warning object Download PDFInfo
- Publication number
- CN106251070A CN106251070A CN201610618345.2A CN201610618345A CN106251070A CN 106251070 A CN106251070 A CN 106251070A CN 201610618345 A CN201610618345 A CN 201610618345A CN 106251070 A CN106251070 A CN 106251070A
- Authority
- CN
- China
- Prior art keywords
- early warning
- main body
- warning
- sample
- user behavior
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06398—Performance of employee with respect to a job function
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The method and device automatically obtaining early warning object that the present invention provides, by obtaining early warning sample main body, obtain the early warning sample main body most like with early warning main body, and the early warning object of the early warning object acquisition early warning main body according to the early warning sample main body most like with early warning main body, solving prior art uses artificial default early warning Object Operations loaded down with trivial details, inefficient technical problem, the early warning object of the early warning sample main body most like with early warning main body is utilized indirectly to obtain the early warning object corresponding with early warning main body dexterously, avoid and manually early warning main body is arranged early warning object, substantially increase the acquisition efficiency obtaining the early warning object corresponding with early warning main body, embody higher intelligent level.
Description
Technical field
The present invention relates to communication technical field, be specifically related to the method and device of a kind of automatic acquisition early warning object.
Background technology
At present, some management software design patterns has warning module, such as employee's image warning module in office management system,
Goods entry, stock and sales forewarning index module sold in forewarning index module and JXC System in marketing system etc..And these are pre-
The process that alert module realizes early warning generally comprises: first preset the warning index corresponding with early warning main body corresponding with warning index
Threshold value of warning and the early warning object corresponding with threshold value of warning, then judge monitoring warning index whether reach early warning threshold
Value, the most then be sent to early warning object by early warning information.
As can be seen here, the early warning object in the warning module of prior art needs manually to pre-set, and uses this people
Work is preset the method for early warning object and be there is complex operation, inefficient problem, therefore needs offer one badly and can automatically obtain early warning pair
The method and device of elephant.
Summary of the invention
The invention provides the method and device of a kind of automatic acquisition early warning object, artificial pre-to solve prior art employing
If early warning Object Operations is loaded down with trivial details, inefficient technical problem.
According to an aspect of the present invention, it is provided that a kind of method of automatic acquisition early warning object, including:
Obtaining early warning sample main body, early warning sample main body is that to be preset with corresponding warning index corresponding with warning index
Threshold value of warning and the main body of the early warning object corresponding with threshold value of warning, wherein the quantity of early warning sample main body be more than one；
Obtain the early warning sample main body most like with early warning main body, as similar early warning main body；
The early warning object of the early warning object acquisition early warning main body according to similar early warning main body.
Further, obtain the early warning sample main body most like with early warning main body to include:
Preset early warning attributes entries；
Gather the user behavior data corresponding with early warning attributes entries of early warning main body, as first user behavioral data；
Gather the user behavior data corresponding with early warning attributes entries of early warning sample main body, as the second user behavior data；
Calculate the similarity between first user behavioral data and the second user behavior data, by corresponding with the similarity of maximum
The early warning sample main body that second user behavior data is corresponding, as the early warning sample main body most like with early warning main body.
Further, the similarity calculated between first user behavioral data and the second user behavior data includes:
Based on first user behavioral data, extract the characteristic vector of early warning main body, as first eigenvector；
Based on the second user behavior data, extract the characteristic vector of early warning sample main body, as second feature vector；
Calculate first eigenvector and the similarity of second feature vector.
Further, include before calculating the similarity of first eigenvector and second feature vector:
First eigenvector and second feature vector are normalized.
Further, include according to the early warning object of the early warning object acquisition early warning main body of similar early warning main body:
Obtain warning index and the threshold value of warning of early warning main body；
The warning object similar early warning main body preset for warning index and threshold value of warning is as the early warning object of early warning main body.
Further, early warning attributes entries includes:
History chat data, job performance, job tenure, task, income level, post classification, Ke Huliang, big customer
One or more combinations in amount, academic title's level entry.
According to a further aspect in the invention, it is provided that the device of a kind of automatic acquisition early warning object, including:
Early warning sample main body acquisition device, is used for obtaining early warning sample main body, and early warning sample main body is corresponding for being preset with
The warning index threshold value of warning corresponding with warning index and the main body of the early warning object corresponding with threshold value of warning, the most in advance
The quantity of alert sample main body is more than one；
Similar early warning main body acquisition device, for obtaining the early warning sample main body most like with early warning main body, as similar early warning
Main body；
Early warning object acquisition device, for the early warning object of the early warning object acquisition early warning main body according to similar early warning main body.
Further, similar early warning main body acquisition device includes:
Early warning attributes entries setting device, is used for presetting early warning attributes entries；
First user behavioral data harvester, for gathering the user behavior number corresponding with early warning attributes entries of early warning main body
According to, as first user behavioral data；
Second user behavior data harvester, for gathering the user row corresponding with early warning attributes entries of early warning sample main body
For data, as the second user behavior data；
Calculate device, for calculating the similarity between first user behavioral data and the second user behavior data, will be with maximum
Early warning sample main body corresponding to the second user behavior data corresponding to similarity, as the early warning sample most like with early warning main body
This main body.
Further, calculate device to include:
First eigenvector acquisition device, for based on first user behavioral data, extracting the characteristic vector of early warning main body, as
First eigenvector；
Second feature vector acquisition device, for based on the second user behavior data, extracting the characteristic vector of early warning sample main body,
As second feature vector；
Similarity Measure device, for calculating first eigenvector and the similarity of second feature vector.
Further, calculate device also to include:
Normalized device, for being normalized first eigenvector and second feature vector.
The method have the advantages that
The method and device automatically obtaining early warning object that the present invention provides, by obtaining early warning sample main body, obtains and early warning
The early warning sample main body that main body is most like, and the early warning object acquisition according to the early warning sample main body most like with early warning main body
The early warning object of early warning main body, solves prior art and uses artificial default early warning Object Operations loaded down with trivial details, and inefficient technology is asked
Topic, utilizes the early warning object of the early warning sample main body most like with early warning main body indirectly to obtain corresponding with early warning main body dexterously
Early warning object, it is to avoid manually early warning main body is arranged early warning object, substantially increases and obtains the early warning corresponding with early warning main body
The acquisition efficiency of object, embodies higher intelligent level.
In addition to objects, features and advantages described above, the present invention also has other objects, features and advantages.
Below with reference to figure, the present invention is further detailed explanation.
Accompanying drawing explanation
The accompanying drawing of the part building the application is used for providing a further understanding of the present invention, and the present invention's is schematic real
Execute example and illustrate for explaining the present invention, not building inappropriate limitation of the present invention.In the accompanying drawings:
Fig. 1 is the method flow diagram obtaining threshold value of warning of the preferred embodiment of the present invention；
Fig. 2 be the preferred embodiment of the present invention for the method flow diagram obtaining threshold value of warning simplifying embodiment；
Fig. 3 is the apparatus structure block diagram obtaining threshold value of warning of the preferred embodiment of the present invention.
Description of reference numerals:
10, early warning sample main body acquisition device；20, similar early warning main body acquisition device；30, early warning object acquisition device.
Detailed description of the invention
Below in conjunction with accompanying drawing, embodiments of the invention are described in detail, but the present invention can be defined by the claims
Implement with the multitude of different ways covered.
With reference to Fig. 1, the preferred embodiments of the present invention provide a kind of method of automatic acquisition early warning object, including:
Step S101, obtains early warning sample main body, and early warning sample main body is to be preset with corresponding warning index and early warning
Threshold value of warning that index is corresponding and the main body of the early warning object corresponding with threshold value of warning, wherein the quantity of early warning sample main body is big
In one；
Step S102, obtains the early warning sample main body most like with early warning main body, as similar early warning main body；
Step S103, according to the early warning object of the early warning object acquisition early warning main body of similar early warning main body.
The method obtaining early warning object that the present invention provides, by obtaining early warning sample main body, obtains with early warning main body
Similar early warning sample main body, and the early warning object acquisition early warning master according to the early warning sample main body most like with early warning main body
The early warning object of body, solves prior art and uses artificial default early warning Object Operations loaded down with trivial details, and inefficient technical problem is ingenious
Ground utilizes the early warning object of the early warning sample main body most like with early warning main body indirectly to obtain the early warning pair corresponding with early warning main body
As, it is to avoid manually early warning main body is arranged early warning object, substantially increase and obtain the early warning object corresponding with early warning main body
Obtain efficiency, embody higher intelligent level.
The early warning main body of the present embodiment indication specifically refers to the threshold value of warning that known warning index is corresponding with warning index,
But the unknown early warning object corresponding with threshold value of warning, and early warning sample main body specifically refers to known warning index and warning index
Corresponding threshold value of warning, and the early warning object corresponding with threshold value of warning.The present embodiment is exactly based on the known early warning object of acquisition
Early warning sample main body, and find the early warning sample main body most like with early warning main body, thus according to early warning main body phase
As the early warning object of early warning sample main body indirectly obtain the early warning object of early warning main body.
The present embodiment passes through to obtain the early warning sample main body most like with early warning main body, and according to the similar early warning master obtained
Body obtains the early warning object of early warning main body, it is provided that a kind of early warning object according to similar main body automatically obtains the side of early warning object
Method, the most efficiently and accurately can obtain the early warning object of early warning main body according to the early warning object of similar early warning main body.Need
Bright, in the present embodiment, the quantity of early warning sample main body is more than one, and in order to improve the standard of the early warning object of early warning main body
Really rate, the quantity of the early warning sample main body that the present embodiment obtains should be big as far as possible, thus in limited early warning sample main body
Obtain the early warning sample main body most like with early warning main body.
Alternatively, obtain the early warning sample main body most like with early warning main body to include:
Preset early warning attributes entries；
Gather the user behavior data corresponding with early warning attributes entries of early warning main body, as first user behavioral data；
Gather the user behavior data corresponding with early warning attributes entries of early warning sample main body, as the second user behavior data；
Calculate the similarity between first user behavioral data and the second user behavior data, by corresponding with the similarity of maximum
The early warning sample main body that second user behavior data is corresponding, as the early warning sample main body most like with early warning main body.
According to daily experience, in the system being provided with warning module, there is the early warning main body of similarity for phase
With warning index and threshold value of warning may corresponding identical early warning object.Such as belong to the early warning main body A of identical post classification
Assume be all sale post classification with early warning main body B(), when for identical warning index (being assumed to be sales volume warning index) and
During threshold value of warning (being assumed to be 100,000), needing the early warning object sent possibility identical (may be all such as to be sent by warning message
To sales manager).And owing to judging that similarity between two main bodys can such as be gone through from different perspectives or dimension is analyzed
History chat data, job performance, job tenure, task, income level, post classification, Ke Huliang, mass clients, academic title
One or more combinations in level entry, therefore the present embodiment is obtaining early warning main body and the user behavior number of early warning sample main body
According to time, first preset early warning attributes entries.
The present embodiment is by presetting early warning attributes entries, and gathers respectively according to early warning attributes entries set in advance
Early warning main body and the user behavior data of early warning sample main body, realize the tolerance to similarity not only by user behavior data,
And by from the similarity between the user behavior data analysis and early warning main body and early warning sample main body of multiple dimensions, have relatively
High reliability, the early warning object obtaining early warning main body for the follow-up early warning object according to similar early warning main body establishes data base
Plinth.
It should be noted that in actual implementation process, the present embodiment is gathering and history chat data early warning attribute
During user behavior data corresponding to entry, often using the descriptor of history chat data or key word as with history chat data
The user behavior data that early warning attributes entries is corresponding.
Alternatively, the similarity calculated between first user behavioral data and the second user behavior data includes:
Based on first user behavioral data, extract the characteristic vector of early warning main body, as first eigenvector；
Based on the second user behavior data, extract the characteristic vector of early warning sample main body, as second feature vector；
Calculate first eigenvector and the similarity of second feature vector.
Similarity in the present embodiment refers to first user behavioral data and similar between the second user behavior data
Degree, can be represented by the distance between respective characteristic of correspondence vector, i.e. first eigenvector and second feature to
Distance between amount.The distance of two characteristic vectors is the shortest, illustrates that the similarity of the two characteristic vector is the biggest.Conventional description
The mode of vector distance has Euclidean distance, cosine angle etc..This is represented by the included angle cosine value calculating two characteristic vectors
The computing formula of the similarity of the user behavior data that characteristic vector is corresponding is:, wherein, X, Y represent respectively feature vector, X and
Characteristic vector Y.Span be [0,1], during closer to 1, then it represents that the similarity between characteristic vector is the highest, otherwise,
Value is closer to 0, then it represents that the similarity between two characteristic vectors is the lowest.Can be relatively by the included angle cosine value calculating characteristic vector
For reflecting the similarity between first user behavioral data and the second user behavior data intuitively.
The present embodiment is by extracting early warning main body and the characteristic vector of early warning sample main body respectively, it is thus achieved that first eigenvector
Vectorial with second feature, and by calculating the similarity between first eigenvector and second feature vector, finally obtain first
Similarity between user behavior data and the second user behavior data.
Alternatively, include before calculating the similarity of first eigenvector and second feature vector:
First eigenvector and second feature vector are normalized.
Due to the first eigenvector extracted respectively according to first user behavioral data and the second user behavior data and
Two characteristic vectors potentially include eigenvalue or quantitative eigenvalue qualitatively, such as post classification early warning attributes entries, the
The data gathered in one user behavior data are for selling post classification, and the data gathered in the second user behavior data are technology hilllock
Position classification, then the present embodiment can by the normalization that the pre-sets rule first eigenvector to obtaining and second feature to
Amount is normalized.Specifically, first post classification quantitative identifying rule is set, then according to this quantitative identifying rule to collection
The user behavior data corresponding with post type early warning attributes entries be normalized.The most such as belong to for task early warning
Property entry, in first user behavioral data gather data be 100,000 sales volumes, in the second user behavior data gather data
Be 300,000 sales volumes, then first the present embodiment can also arrange task normalization rule, and normalization scope is arranged on
Between 0-100, referring in particular to table 1.In like manner, for the user behavior data that other early warning attributes entries are corresponding, the present embodiment is also
By the normalization rule pre-set, it can be normalized, thus facilitate follow-up accurate calculating similarity.
Table 1
Alternatively, include according to the early warning object of the early warning object acquisition early warning main body of similar early warning main body:
Obtain warning index and the threshold value of warning of early warning main body；
The warning object similar early warning main body preset for warning index and threshold value of warning is as the early warning object of early warning main body.
Specifically, after the present embodiment obtains the early warning sample main body most like with early warning main body, just can according to early warning master
The early warning object of the early warning sample main body that body is most like obtains early warning object.Due to for each early warning sample main body, early warning
Module generally comprise multiple warning index (in such as office management system employee's image warning module include employee ask virtual degree,
Employee asks for leave natural law and employee stays away from work without leave number of times warning index etc.), and for each warning index, it may include again many
Individual threshold value of warning (is such as asked for leave natural law warning index for employee, is arranged employee's natural law of asking for leave and sent early warning more than 2 days to supervisor
Information, sent early warning information more than 3 days to manager, sent early warning information etc. more than 5 days to general manager) etc., therefore the present embodiment exists
During the early warning object of the early warning object acquisition early warning main body according to similar early warning main body, first obtain the warning index of early warning main body
And threshold value of warning, then using similar early warning main body for the warning object of above-mentioned warning index and threshold value of warning as early warning main body
Early warning object.
Alternatively, early warning attributes entries includes:
History chat data, job performance, job tenure, task, income level, post classification, Ke Huliang, big customer
One or more combinations in amount, academic title's level entry.
Early warning attributes entries in the present embodiment is not limited to history chat data, job performance, job tenure, work are appointed
In business, income level, post classification, Ke Huliang, mass clients, academic title's level entry one or more combination, specifically by with
Family is self-defined.
Simplify embodiment below for one the method automatically obtaining early warning object of the present invention is carried out furtherly
Bright.
With reference to Fig. 2, the method automatically obtaining early warning object in the present embodiment includes:
Step S201, obtains early warning sample main body, and early warning sample main body is to be preset with corresponding warning index and early warning
Threshold value of warning that index is corresponding and the main body of the early warning object corresponding with threshold value of warning, wherein the quantity of early warning sample main body is big
In one.
Specifically, it is assumed that the early warning sample main body of the present embodiment acquisition is 500, and known and each early warning sample master
Warning index that body is corresponding and the early warning object corresponding with warning index.It should be noted that the early warning that the present embodiment obtains
The quantity of sample main body should be big as far as possible, thus obtains most like with early warning main body pre-in limited early warning sample main body
Alert sample main body.
Step S202, presets early warning attributes entries.
Specifically, the present embodiment early warning set in advance attributes entries can be history chat data, job performance, work
One or more groups in the term of office, task, income level, post classification, Ke Huliang, mass clients, academic title's level entry
Close.But it is also not necessarily limited to choose history chat data, job performance, job tenure, task, income level, post classification, visitor
One or more combinations in family amount, mass clients, academic title's level entry, can carry out self-defined setting as required.Assume
The early warning attributes entries that the present embodiment pre-sets includes three, and specially job tenure, task, post classification early warning belongs to
Property entry.
Step S203, gathers the user behavior data corresponding with early warning attributes entries of early warning main body, as first user
Behavioral data.
Specifically, it is assumed that the present embodiment collects for early warning main body A and divides with three the early warning attributes entries pre-set
Not corresponding user behavior data is: job tenure early warning attributes entries-5 years；Task early warning attributes entries-30 ten thousand is sold
Volume；Post classification early warning attributes entries: sell classification.
Step S204, gathers the user behavior data corresponding with early warning attributes entries of early warning sample main body, as second
User behavior data.
Specifically, the present embodiment respectively to 500 the early warning sample main bodys obtained in step S201 with pre-set
User behavior data corresponding to three early warning attributes entries is acquired.
Step S205, calculates the similarity between first user behavioral data and the second user behavior data, will be with maximum
Early warning sample main body corresponding to the second user behavior data corresponding to similarity, as the early warning sample most like with early warning main body
This main body.
Specifically, when the present embodiment calculates the similarity between first user behavioral data and the second user behavior data,
It is primarily based on first user behavioral data, extracts the characteristic vector of early warning main body, as first eigenvector, and based on second
User behavior data, extracts the characteristic vector of early warning sample main body, as second feature vector, then calculates first eigenvector
Similarity with second feature vector.It should be noted that the present embodiment is after extracting the first and second characteristic vectors, calculate
Also include first eigenvector and second feature vector are carried out before the similarity of first eigenvector and second feature vector
Normalized, specifically can refer to the normalization rule in table 1.
Additionally, due to the early warning sample main body that the present embodiment obtains is 500, then calculating the first and second user behaviors
During similarity between data, first user behavioral data is used with 500 early warning sample main bodys each self-corresponding second respectively
Family behavioral data carries out Similarity Measure, then by early warning corresponding for second user behavior data corresponding with maximum similarity
Sample main body, as the early warning sample main body most like with early warning main body, it is assumed that obtain in 500 early warning sample main bodys with pre-
The early warning sample main body warning main body most like is early warning sample main body B.
Step S206, obtains warning index and the threshold value of warning of early warning main body.
Specifically, it is assumed that the present embodiment obtain the warning index corresponding with early warning main body A be sales volume warning index, and with
The threshold value of warning that this warning index is corresponding is 100,000.
Step S207, the warning object similar early warning main body preset for warning index and threshold value of warning is as early warning master
The early warning object of body.
Specifically, the similar early warning main body obtained due to the present embodiment is early warning sample main body B, then can be by early warning sample master
Body B for the warning object of the warning index obtained in step S206 and threshold value of warning as the early warning object of early warning main body A.Also
Will the known warning index arranged for early warning sample main body B be sales volume warning index, corresponding threshold value of warning
It it is the early warning object as early warning main body A of early warning object when 100,000.
As can be seen here, the present embodiment, by obtaining early warning sample main body, obtains the early warning sample most like with early warning main body
Main body, and can automatically obtain the pre-of early warning main body according to the early warning object of the early warning sample main body most like with early warning main body
Alert object, solves prior art and uses artificial default early warning Object Operations loaded down with trivial details, and inefficient technical problem utilizes dexterously
The early warning object of the early warning sample main body most like with early warning main body obtains the early warning object corresponding with early warning main body indirectly, it is to avoid
Manually early warning main body is arranged early warning object, substantially increase and obtain the acquisition of the early warning object corresponding with early warning main body and imitate
Rate, embodies higher intelligent level.
With reference to Fig. 3, the device automatically obtaining early warning object that the preferred embodiments of the present invention provide, including:
Early warning sample main body acquisition device 10, is used for obtaining early warning sample main body, and early warning sample main body is right with it for being preset with
Threshold value of warning that the warning index answered is corresponding with warning index and the main body of the early warning object corresponding with threshold value of warning, wherein
The quantity of early warning sample main body is more than one；
Similar early warning main body acquisition device 20, for obtaining the early warning sample main body most like with early warning main body, as similar pre-
Alert main body；
Early warning object acquisition device 30, for the early warning object of the early warning object acquisition early warning main body according to similar early warning main body.
Alternatively, similar early warning main body acquisition device 20 includes:
Early warning attributes entries setting device, is used for presetting early warning attributes entries；
First user behavioral data harvester, for gathering the user behavior number corresponding with early warning attributes entries of early warning main body
According to, as first user behavioral data；
Second user behavior data harvester, for gathering the user row corresponding with early warning attributes entries of early warning sample main body
For data, as the second user behavior data；
Calculate device, for calculating the similarity between first user behavioral data and the second user behavior data, will be with maximum
Early warning sample main body corresponding to the second user behavior data corresponding to similarity, as the early warning sample most like with early warning main body
This main body.
Alternatively, calculate device to include:
First eigenvector acquisition device, for based on first user behavioral data, extracting the characteristic vector of early warning main body, as
First eigenvector；
Second feature vector acquisition device, for based on the second user behavior data, extracting the characteristic vector of early warning sample main body,
As second feature vector；
Similarity Measure device, for calculating first eigenvector and the similarity of second feature vector.
Alternatively, calculate device also to include:
Normalized device, for being normalized first eigenvector and second feature vector.
The device obtaining early warning object that the present invention provides, by obtaining early warning sample main body, obtains with early warning main body
Similar early warning sample main body, and the early warning object acquisition early warning master according to the early warning sample main body most like with early warning main body
The early warning object of body, solves prior art and uses artificial default early warning Object Operations loaded down with trivial details, and inefficient technical problem is ingenious
Ground utilizes the early warning object of the early warning sample main body most like with early warning main body indirectly to obtain the early warning pair corresponding with early warning main body
As, it is to avoid manually early warning main body is arranged early warning object, substantially increase and obtain the early warning object corresponding with early warning main body
Obtain efficiency, embody higher intelligent level.
Specific works process and the operation principle of the device of the present embodiment acquisition early warning object can refer to obtaining of the present embodiment
Take work process and the operation principle of the method for early warning object.
These are only the preferred embodiments of the present invention, be not limited to the present invention, for those skilled in the art
For Yuan, the present invention can have various modifications and variations.All within the spirit and principles in the present invention, any amendment of being made,
Equivalent, improvement etc., should be included within the scope of the present invention.
Claims (10)
1. the method for an automatic acquisition early warning object, it is characterised in that including:
Obtaining early warning sample main body, described early warning sample main body is to be preset with corresponding warning index and described early warning
Threshold value of warning that index is corresponding and the main body of the early warning object corresponding with described threshold value of warning, wherein said early warning sample main body
Quantity more than one；
Obtain the early warning sample main body most like with described early warning main body, as similar early warning main body；
The early warning object of early warning main body described in early warning object acquisition according to described similar early warning main body.
The method of automatic acquisition early warning object the most according to claim 1, it is characterised in that obtain and described early warning main body
Most like early warning sample main body includes:
Preset early warning attributes entries；
Gather the user behavior data corresponding with described early warning attributes entries of described early warning main body, as first user behavior number
According to；
Gather the user behavior data corresponding with described early warning attributes entries of described early warning sample main body, as second user's row
For data；
Calculate the similarity between described first user behavioral data and described second user behavior data, by similar to maximum
Spend the early warning sample main body that the second corresponding user behavior data is corresponding, as the early warning sample most like with described early warning main body
Main body.
The method of automatic acquisition early warning object the most according to claim 2, it is characterised in that calculate described first user row
Include for the similarity between data and described second user behavior data:
Based on described first user behavioral data, extract the characteristic vector of described early warning main body, as first eigenvector；
Based on described second user behavior data, extract the characteristic vector of described early warning sample main body, as second feature vector；
Calculate described first eigenvector and the similarity of described second feature vector.
The method of automatic acquisition early warning object the most according to claim 3, it is characterised in that calculate described fisrt feature to
Include before the similarity of amount and described second feature vector:
Described first eigenvector and described second feature vector are normalized.
The method of automatic acquisition early warning object the most according to claim 4, it is characterised in that according to described similar early warning master
Described in the early warning object acquisition of body, the early warning object of early warning main body includes:
Obtain warning index and the threshold value of warning of described early warning main body；
The warning object described similar early warning main body preset for described warning index and threshold value of warning is as described early warning master
The early warning object of body.
The method of automatic acquisition early warning object the most according to claim 5, it is characterised in that described early warning attributes entries bag
Include:
History chat data, job performance, job tenure, task, income level, post classification, Ke Huliang, big customer
One or more combinations in amount, academic title's level entry.
7. the device of an automatic acquisition early warning object, it is characterised in that including:
Early warning sample main body acquisition device, is used for obtaining early warning sample main body, and described early warning sample main body is for being preset with and it
Threshold value of warning that corresponding warning index is corresponding with described warning index and the early warning object corresponding with described threshold value of warning
Main body, the quantity of wherein said early warning sample main body is more than one；
Similar early warning main body acquisition device, for obtaining the early warning sample main body most like with described early warning main body, as similar
Early warning main body；
Early warning object acquisition device, for the early warning according to early warning main body described in the early warning object acquisition of described similar early warning main body
Object.
The device of automatic acquisition early warning object the most according to claim 7, it is characterised in that described similar early warning main body obtains
Fetching is put and is included:
Early warning attributes entries setting device, is used for presetting early warning attributes entries；
First user behavioral data harvester, for gathering the use corresponding with described early warning attributes entries of described early warning main body
Family behavioral data, as first user behavioral data；
Second user behavior data harvester, for gathering the corresponding with described early warning attributes entries of described early warning sample main body
User behavior data, as the second user behavior data；
Calculate device, for calculating the similarity between described first user behavioral data and described second user behavior data,
By early warning sample main body corresponding for second user behavior data corresponding with maximum similarity, as with described early warning main body
Similar early warning sample main body.
The device of automatic acquisition early warning object the most according to claim 8, it is characterised in that described calculating device includes:
First eigenvector acquisition device, for based on described first user behavioral data, extracts the feature of described early warning main body
Vector, as first eigenvector；
Second feature vector acquisition device, for based on described second user behavior data, extracts described early warning sample main body
Characteristic vector, as second feature vector；
Similarity Measure device, for calculating described first eigenvector and the similarity of described second feature vector.
The device of automatic acquisition early warning object the most according to claim 9, it is characterised in that described calculating device also wraps
Include:
Normalized device, for being normalized described first eigenvector and described second feature vector.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610618345.2A CN106251070A (en) | 2016-08-01 | 2016-08-01 | A kind of method and device of automatic acquisition early warning object |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610618345.2A CN106251070A (en) | 2016-08-01 | 2016-08-01 | A kind of method and device of automatic acquisition early warning object |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106251070A true CN106251070A (en) | 2016-12-21 |
Family
ID=57605617
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610618345.2A Pending CN106251070A (en) | 2016-08-01 | 2016-08-01 | A kind of method and device of automatic acquisition early warning object |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106251070A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103226651A (en) * | 2013-03-23 | 2013-07-31 | 中国水利电力物资有限公司 | Wind turbine state evaluation and early-warning method and system based on similarity statistics |
| EP2688015A1 (en) * | 2012-07-20 | 2014-01-22 | Tata Consultancy Services Limited | Method and system for adaptive forecast of energy resources |
| CN103854104A (en) * | 2012-11-29 | 2014-06-11 | 瀚科科技（大连）有限公司 | Enterprise-employee certificate early-warning system |
-
2016
- 2016-08-01 CN CN201610618345.2A patent/CN106251070A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2688015A1 (en) * | 2012-07-20 | 2014-01-22 | Tata Consultancy Services Limited | Method and system for adaptive forecast of energy resources |
| CN103854104A (en) * | 2012-11-29 | 2014-06-11 | 瀚科科技（大连）有限公司 | Enterprise-employee certificate early-warning system |
| CN103226651A (en) * | 2013-03-23 | 2013-07-31 | 中国水利电力物资有限公司 | Wind turbine state evaluation and early-warning method and system based on similarity statistics |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110765770B (en) | Automatic contract generation method and device | |
| CN109598509A (en) | The recognition methods of risk clique and device | |
| CN105404627B (en) | It is a kind of for determining the method and apparatus of search result | |
| CN110363076A (en) | Personal information correlating method, device and terminal device | |
| CN107193894A (en) | Data processing method, individual discrimination method and relevant apparatus | |
| CN107563757A (en) | The method and device of data risk control | |
| CN106843941B (en) | Information processing method, device and computer equipment | |
| CN110648172B (en) | Identity recognition method and system integrating multiple mobile devices | |
| CN114742477B (en) | Enterprise order data processing method, device, equipment and storage medium | |
| CN106095939B (en) | The acquisition methods and device of account authority | |
| CN105376223B (en) | The reliability degree calculation method of network identity relationship | |
| CN104517052B (en) | Invasion detection method and device | |
| CN109213921A (en) | A kind of searching method and device of merchandise news | |
| CN113157947A (en) | Knowledge graph construction method, tool, device and server | |
| CN109615172A (en) | A kind of method and terminal handling examination data | |
| CN111177208A (en) | Power consumption abnormity detection method based on big data analysis | |
| CN108921656A (en) | Business recommended method and device | |
| CN106886517A (en) | Business site selecting method, device and system | |
| CN110377977A (en) | Sensitive information leakage detection method and device and storage medium | |
| CN110909129B (en) | Abnormal complaint event identification method and device | |
| CN108664605A (en) | A kind of model evaluation method and system | |
| CN109558462A (en) | Data statistical approach and device | |
| CN111985577A (en) | Customer value labeling method, device, equipment and medium based on artificial intelligence | |
| CN116188050A (en) | Takeaway platform information processing system based on data analysis | |
| CN111897898A (en) | Configuration method and device for position of Internet of things equipment and computer equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161221 |
|
| RJ01 | Rejection of invention patent application after publication |