CN110503249A - One kind complaining prediction technique caused by having a power failure - Google Patents
One kind complaining prediction technique caused by having a power failure Download PDFInfo
- Publication number
- CN110503249A CN110503249A CN201910723742.XA CN201910723742A CN110503249A CN 110503249 A CN110503249 A CN 110503249A CN 201910723742 A CN201910723742 A CN 201910723742A CN 110503249 A CN110503249 A CN 110503249A
- Authority
- CN
- China
- Prior art keywords
- power failure
- traffic data
- caused
- data
- outage
- 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
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000035945 sensitivity Effects 0.000 claims abstract description 12
- 238000005303 weighing Methods 0.000 claims abstract description 4
- 230000009467 reduction Effects 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 238000003066 decision tree Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 5
- 238000007477 logistic regression Methods 0.000 claims description 3
- 238000010801 machine learning Methods 0.000 claims description 3
- 238000000513 principal component analysis Methods 0.000 claims description 3
- 238000007637 random forest analysis Methods 0.000 claims description 3
- 238000012706 support-vector machine Methods 0.000 claims description 3
- 235000013399 edible fruits Nutrition 0.000 claims 1
- 238000003780 insertion Methods 0.000 claims 1
- 230000037431 insertion Effects 0.000 claims 1
- 238000013459 approach Methods 0.000 abstract description 4
- 238000004364 calculation method Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 5
- 238000010276 construction Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 230000005611 electricity Effects 0.000 description 3
- 230000008451 emotion Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000007405 data analysis Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 208000025174 PANDAS Diseases 0.000 description 1
- 208000021155 Paediatric autoimmune neuropsychiatric disorders associated with streptococcal infection Diseases 0.000 description 1
- 240000000220 Panda oleosa Species 0.000 description 1
- 235000016496 Panda oleosa Nutrition 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Public Health (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to one kind, and prediction technique is complained caused by having a power failure, pass through each dimension data of outage information, client's account information and the traffic information acquisition of user in collection national grid PMS2.0 system, marketing system, 95598 customer service systems, and numbered by subscriber board area with number being associated in every family to the data in three systems, and thus distinguish the traffic data of outage traffic data and non-outage；For the traffic data of outage; power failure duration first is determined with Information Entropy, whether is given advance notice, the weight of the influential feature of peak power off time accounting traffic data; afterwards according to its weighing factor; power-off event is shielded to power failure traffic data bring difference between platform area, is portrayed on this basis according to daily traffic data and power failure traffic data come the sensitivity of power consumer.The present invention expands the range of prediction, alert discriminant approach more closing to reality.
Description
Technical field
The present invention relates to information technology field technical fields, and in particular to a kind of that prediction technique is complained caused by having a power failure.
Background technique
Currently, accounting for the large percentage of customer complaint due to complaining caused by frequently having a power failure, Ji north limited power is netted according to state
The complaint data analysis that company provides, the complaint of frequent power failure classification have accounted for complaining 40 or so the percent of total amount, individually
Time has even accounted for complaining 50 the percent of total amount.Therefore document ([1] Xu Xin, Wang Li, Sun Zhijie, Gong Dongmei, Zhang Ling
Space, a kind of frequent power failure complaint Early-warning Model [J] information-recording material based on data mining of Liu Xiaowei, Qin Fengyuan, 2017,
18 (02): 64-66.) describe a kind of for the complaint Early-warning Model frequently having a power failure, same user in two months is provided in text
The power failure occurred three times or more is then considered as frequent power failure, and by Chinese Word Automatic Segmentation, from the troublshooting letter generated
Address information and power off time are excavated in breath and outage information to count the frequency of power cut of certain power failure unit, and combine the exhibition of cloud map
Existing warning information forms early warning mechanism in advance, complains early warning to achieve the purpose that frequently to have a power failure.
Frequent power failure data and address information in document are all from work order content, since work order data are limited and divide
The actual effect of word algorithm is filled in specification by work order and is affected, and can not achieve the accurate of frequent power failure data and address information
Statistics；It proposes to make complaint early warning to occurring having a power failure three times or more in two months in text, but in fact, is directed to the complaint of user
Early warning should be using the practical impression of user as target, and modeling process does not capture user experience from multi-angle, excavates production in text
The raw potential cause complained；Due to caused by frequently having a power failure complain and it is not all embodied in work order, part frequently have a power failure causes
The discontented generation that will cause negative emotions of client and additional customer service (such as: paying a return visit, explain), so that part be caused frequently to stop
The complaint that, explanation poor to customer service attitude is not approved can be converted into work order by complaining caused by electric, cause data to hold accurate.
Summary of the invention
The purpose of the present invention is to provide a kind of range for expanding prediction, alert discriminant approach more closing to reality, from
95598 traffic data of angle analysis of family experience is compared to the tendency that can more reflect customer complaint with the frequent definition that has a power failure by stopping
Prediction technique is complained caused by electric.
Complaint actually more than half is all related to power-off event, also belongs to power failure to the complaint frequently having a power failure and causes to throw
The a part told, having a power failure, it is excessively frequent to arrange, and seriously affects user power utilization experience；Scheduled outage has a power failure in advance, be delayed power transmission, shadow
Ring user's living arrangement；Unplanned power off time is too long, and without accurate electric power feeding time response etc..It uses due to the above reasons,
Family can decline the degree of recognition of power supply, while generate negative emotions, generate to complain or increase and complain in normal clients service
Probability.
Therefore mass data should be collected to the complaint of electric service prediction, the relevant number that has a power failure is obtained by data correlation
The traffic of power-off event is fed back according to, the information of user and user, it is not comprehensively that analysis is carried out only by work order data
And inaccuracy；Call service is the major way that grid company carries out customer service, therefore by different anti-with the traffic of group
User can be excavated to the sensitivity of power-off event, due to different user pair by answering and (complaining, report for repairment, pressing, finishing, seeking advice from)
It is different in the degree of understanding of power-off event, therefore sensitivity is also different, this is extremely important in complaining prediction；Only to frequently stopping
Electricity carry out early warning be it is not comprehensive enough, early warning should be carried out to all stop caused by power transmission complaint, and will as stopping caused by power transmission but
Directly the complaint carried out that has a power failure is not taken into account.
The main object of the present invention is pre- to all kinds of complaints progress caused by power transmission are stopped；Secondly, needing in this process
The model of user's power failure sensitivity is established as one of the feature for complaining analysis user behavior in prediction；Again, to traffic number
Traffic intensive period (outage) and daily traffic period are distinguished when being analyzed according to feature, and are comprehensively considered, as
User's susceptibility describes basis.
Technical solution of the present invention:
The present invention is by collecting national grid PMS2.0 system, marketing system, the outage information in 95598 customer service systems, client
The traffic information acquisition of account information and user multi-dimensional data, and numbered by subscriber board area and with number to three being in every family
Data in system are associated with, and have thus distinguished the traffic data of outage traffic data and non-outage.For
The traffic data of outage, since different power-off event attributes are different, identical for susceptibility area also can be in traffic number
According to it is upper occur it is inconsistent, therefore first with Information Entropy determine power failure duration, whether give advance notice, peak power off time accounting etc. is to traffic
The weight of the influential feature of data, after according to its weighing factor difference, shield different power-off events between platform area have a power failure talk about
Business data bring difference, on this basis according to daily traffic data and power failure traffic data come the sensitivity of power consumer
It is portrayed.
Since traffic data feature quantity is more, needs to carry out dimensionality reduction to data, compare Principal Component Analysis
(Principal Components Analysis, PCA) and t- distributed random neighborhood are embedded in (t-distributed
Stochastic Neighbor Embedding, t-SNE) actual effect, choose t-SNE traffic data carry out dimensionality reduction.It is right
Data after dimensionality reduction carry out clustering, and delimit sensitivity grade.By being found compared with k mean algorithm (K-means),
The cluster result of gauss hybrid models (Gaussian Mixture Model, GMM) is more in line with reality, and to the sensitivity of user
Degree is divided into 5 grades.Described 5 grades respectively sensitive, more sensitive, normal, less sensitive, insensitive.
On the basis of client-aware degree, it can be directed to the power failure of single, according to power failure duration, whether belong to and frequently stop
Electricity, stops the features such as radio area and its susceptibility at power off time section, and unplanned repairing is had a power failure and counted by the way of machine learning
Draw complains probability to be predicted caused by having a power failure.Wherein, it is found by statistics, related complain caused by having a power failure mostly occurs in
In 12 hours after outage and telegram in reply, therefore the complaint of the affected area intra domain user after sending a telegram in reply in 12 hours also belongs to power failure
Caused complaint.According to the prediction model assessment of comparison decision tree, support vector machines and logistic regression, decision Tree algorithms result is more
It is excellent, it is more suitable for complaining caused by having a power failure and predicts.And improved using random forests algorithm, effectively reduce prediction error
Probability.
Beneficial effects of the present invention:
Compared to existing calculation method, firstly, expanding the range of prediction, customer electricity is experienced in influence according to having a power failure
It is contacting, for having a power failure, caused all kinds of complaints are predicted；The accuracy of data and comprehensive there are promotion, grid company
Integrality and accuracy after internal three inter-system datas association will be apparently higher than the result of Chinese word segmentation processing work order content；
Early warning discriminant approach more closing to reality, proposes the calculation method of client-aware degree, from the angle analysis 95598 of user experience
Traffic data is compared to the tendency that can more reflect customer complaint with the frequent definition that has a power failure.
Detailed description of the invention
Fig. 1 is the association schematic diagram of present system data；
Fig. 2-1 is t-SNE dimensionality reduction result figure of the present invention；
Fig. 2-2 is CAP dimensionality reduction result figure of the present invention；
Fig. 3-1 is GMM cluster result comparison diagram of the present invention；
Fig. 3-2 is k-means cluster result comparison diagram of the present invention；
Fig. 4 is complaint time interval distribution map after present invention telegram in reply.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of embodiments of the present application, instead of all the embodiments.Below
Description only actually at least one exemplary embodiment be it is illustrative, never as to the application and its application or making
Any restrictions.Based on the embodiment in the application, those of ordinary skill in the art are not making creative work premise
Under every other embodiment obtained, shall fall in the protection scope of this application.
It should be noted that term used herein above is merely to describe specific embodiment, and be not intended to restricted root
According to the illustrative embodiments of the application.As used herein, unless the context clearly indicates otherwise, otherwise singular
Also it is intended to include plural form, additionally, it should be understood that, when in the present specification using term "comprising" and/or " packet
Include " when, indicate existing characteristics, step, operation, device, component and/or their combination.
Unless specifically stated otherwise, positioned opposite, the digital table of the component and step that otherwise illustrate in these embodiments
Up to the unlimited scope of the present application processed of formula and numerical value.Simultaneously, it should be appreciated that for ease of description, each portion shown in attached drawing
The size divided not is to draw according to actual proportionate relationship.For technology, side known to person of ordinary skill in the relevant
Method and equipment may be not discussed in detail, but in the appropriate case, and the technology, method and apparatus should be considered as authorizing explanation
A part of book.In shown here and discussion all examples, any occurrence should be construed as merely illustratively, and
Not by way of limitation.Therefore, the other examples of exemplary embodiment can have different values.It should also be noted that similar label
Similar terms are indicated in following attached drawing with letter, therefore, once it is defined in a certain Xiang Yi attached drawing, then subsequent attached
It does not need that it is further discussed in figure.
In the description of the present application, it is to be understood that the noun of locality such as " front, rear, top, and bottom, left and right ", " laterally, vertical,
Vertically, orientation or positional relationship indicated by level " and " top, bottom " etc. is normally based on orientation or position shown in the drawings and closes
System is merely for convenience of description the application and simplifies description, and in the absence of explanation to the contrary, these nouns of locality do not indicate that
It must have a particular orientation or be constructed and operated in a specific orientation with the device or element for implying signified, therefore cannot manage
Solution is the limitation to the application protection scope；The noun of locality " inside and outside " refers to inside and outside the profile relative to each component itself.
For ease of description, spatially relative term can be used herein, as " ... on ", " ... top ",
" ... upper surface ", " above " etc., for describing such as a device shown in the figure or feature and other devices or spy
The spatial relation of sign.It should be understood that spatially relative term is intended to comprising the orientation in addition to device described in figure
Except different direction in use or operation.For example, being described as if the device in attached drawing is squeezed " in other devices
It will be positioned as " under other devices or construction after part or construction top " or the device of " on other devices or construction "
Side " or " under other devices or construction ".Thus, exemplary term " ... top " may include " ... top " and
" in ... lower section " two kinds of orientation.The device can also be positioned with other different modes and (is rotated by 90 ° or in other orientation), and
And respective explanations are made to the opposite description in space used herein above.
In addition, it should be noted that, limiting components using the words such as " first ", " second ", it is only for be convenient for
Corresponding components are distinguished, do not have Stated otherwise such as, there is no particular meanings for above-mentioned word, therefore should not be understood as to this
Apply for the limitation of protection scope.
Technical solution of the present invention, structure are described in further detail below in conjunction with attached drawing.
Instance data source:
Baoding Utilities Electric Co., national grid Hebei Electric Power Corporation PMS2.0 system, marketing system, 95598 customer service systems
Data software for calculation and kit:
Python3.7.1(pymysql, numpy, math, matplotlib, pandas, sklearn), PyCharm
2018.3.2
Database:
Database、Navicat Premium 12
Such as attached drawing 1, by the data pick-up in three systems into MySQL database, by the not area homologous ray Zhong Tai number and
With number the traffic feedback of the user information and user of the details of power failure and influence associating in every family, it is convenient for
Subsequent data screening and Feature Selection.
After the outage traffic data in area is not separated with daily traffic data on the same stage, the traffic data of outage by
To the influence of different power-off events, the sensitivity difference between platform area cannot be objectively showed.The length of time of power failure, having a power failure is
It is no give advance notice, whether belong to frequently power failure, the influence user experience that the peak of power consumption time accounting among power off time is
Feature, therefore be determined that this four features see attached list 1 to the weighing factor of outage traffic data with Information Entropy, accordingly can be with
Eliminate outage traffic data influences as caused by power failure attribute difference, realizes power failure data normalization.
On this basis, more by standardized power failure traffic data and daily traffic data characteristic quantity, using the side t-SNE
Method carries out dimensionality reduction to 2 dimensions to data, and as shown in attached drawing 2-1 and Fig. 2-2, dimensionality reduction effect is more preferable compared with Principal Component Analysis, data
Feature retains more complete in two-dimensional space.The sensitivity of client is clustered according to the 2 of traffic data dimension dimensionality reduction results afterwards
Classification.
As shown in attached drawing 3-1 and Fig. 3-2, K-means method is entirely to be completed according to distance to the cluster of data point, is not had
Consider higher-dimension lead data characteristics lower dimensional space mapping result characteristic aggregation as a result, the comparatively cluster of gauss hybrid models
As a result comparatively more meet reality, the parametric solution method of dimensional gaussian distribution is desired maximum calculated in gauss hybrid models
Method.
Client's power failure susceptibility is the high information density data set that low information density is integrated, and is based on client-aware degree,
More data informations can be expressed with less characteristic quantity during prediction has a power failure and causes customer complaint, make prediction result
It is more in line with reality, characteristic data set is as shown in attached drawing 2-1 and 2-2.Cause the Annual distribution complained statistics such as according to by power failure
Attached drawing 4, the complaint after discovery is sent a telegram in reply concentrated in preceding 11 hours, were found by calculating, and the complaint after telegram in reply in 12 hours accounts for
The 79.68% of total amount is complained after telegram in reply, therefore whether occurs customer complaint using in after outage and telegram in reply 12 hours as label,
Complaint after power failure in 12 hours is considered as by the recessive complaint caused that has a power failure.
Complaint is predicted using support vector machines, decision tree and logistic regression, 10 foldings intersection is splitted data into and tests
Card form, and the results are shown in Table 3 using accuracy rate (Precision), recall rate (Recall) and F1 value comparison algorithms of different, certainly
The prediction result of plan tree is substantially better than other two kinds of algorithms.The method that Gu Xuanzejueceshu is complained as prediction power failure, and use
Random forests algorithm realizes parallel integrated learning approach, comprehensively considers to the classification results of multiple decision trees, reduces erroneous judgement
Probability.
1 weight definitive result of table
2 customer complaint predicted characteristics of table
3 machine learning algorithm evaluation result of table
In the present invention: abbreviation and Key Term definition
Customer service: customer service
Telegram in reply: restore power transmission after power failure
Platform area: whole loads of a 10kV transformer back segment on substation feeder
Power failure susceptibility: after power failure there are discontented and negative emotions probability to the sensitivity of power-off event in client.
The present invention is complete data processing and calculation method, including multi-source data association, traffic data standardization and dimensionality reduction
Processing, user's power failure susceptibility model, to the caused complaint prediction etc. that has a power failure, and the selection of calculation method and model is according to reality
What effect was determined, uniqueness is not had, if this programme carries out data analysis and modeling for other field, part is counted
Calculation method, which is replaced, may obtain better result.
Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although
Present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: it still may be used
To modify to technical solution documented by previous embodiment or equivalent replacement of some of the technical features；Make
It is obvious for being combined for those skilled in the art to multiple technical solutions of the invention.And these are modified or replace
It changes, the spirit and scope for technical solution of the embodiment of the present invention that it does not separate the essence of the corresponding technical solution.
Claims (10)
1. a kind of complain prediction technique caused by having a power failure, which comprises the following steps:
(1) by collecting national grid PMS2.0 system, marketing system, the outage information in 95598 customer service systems, client's account
The each dimension data of the traffic information acquisition of information and user, and by subscriber board area number and use in every family number to three systems
Interior data are associated with, and have thus distinguished the traffic data of outage traffic data and non-outage；
(2) for the traffic data of outage, first with Information Entropy determine power failure duration, whether give advance notice, peak have a power failure when
Between the influential feature of accounting traffic data weight, after according to its weighing factor, shield power-off event and stop between platform area
Phone business data bring difference, on this basis according to daily traffic data and power failure traffic data come the sensitivity of power consumer
Degree is portrayed.
A kind of prediction technique is complained caused by having a power failure 2. according to claim 1, which is characterized in that traffic data into
Row dimensionality reduction, compare Principal Component Analysis and t- distributed random neighborhood insertion actual effect, choose t-SNE traffic data into
Row dimensionality reduction.
3. a kind of complaint prediction technique caused by having a power failure according to claim 2, which is characterized in that the number after dimensionality reduction
According to progress clustering, and delimit sensitivity grade.
A kind of prediction technique is complained caused by having a power failure 4. according to claim 3, which is characterized in that by with k mean value
Algorithm K-means compares, and gauss hybrid models is selected, using the cluster result of gauss hybrid models, to the sensitive journey of user
Degree is divided into 5 grades.
5. a kind of complaint prediction technique caused by having a power failure according to claim 4, which is characterized in that 5 grades of difference
It is sensitive, more sensitive, normal, less sensitive, insensitive.
6. a kind of complaint prediction technique caused by having a power failure according to claim 1, which is characterized in that in client-aware degree
On the basis of, for the power failure of single, according to power failure duration, whether belong to frequent power failure, power off time section, stop radio area and its
The features such as susceptibility carry out complaint probability caused by unplanned repairing power failure and scheduled outage by the way of machine learning pre-
It surveys.
7. a kind of complaint prediction technique caused by having a power failure according to claim 1, which is characterized in that 12 hours after telegram in reply
The complaint of interior affected area intra domain user belongs to complains caused by power failure.
8. a kind of complaint prediction technique caused by having a power failure according to claim 7, which is characterized in that prediction model assessment
Model used is the prediction model for comparing decision tree, support vector machines and logistic regression.
9. a kind of complaint prediction technique caused by having a power failure according to claim 8, which is characterized in that decision Tree algorithms knot
Fruit predicts caused complain that have a power failure.
10. a kind of complaint prediction technique caused by having a power failure according to claim 9, which is characterized in that preference pattern is pre-
After survey, improved using random forests algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910723742.XA CN110503249A (en) | 2019-08-07 | 2019-08-07 | One kind complaining prediction technique caused by having a power failure |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910723742.XA CN110503249A (en) | 2019-08-07 | 2019-08-07 | One kind complaining prediction technique caused by having a power failure |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110503249A true CN110503249A (en) | 2019-11-26 |
Family
ID=68587966
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910723742.XA Pending CN110503249A (en) | 2019-08-07 | 2019-08-07 | One kind complaining prediction technique caused by having a power failure |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110503249A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111353792A (en) * | 2020-05-25 | 2020-06-30 | 广东电网有限责任公司惠州供电局 | Client portrait system with visual display and data analysis functions |
| CN111553816A (en) * | 2020-04-20 | 2020-08-18 | 北京北大软件工程股份有限公司 | Method and device for analyzing administrative review influence factors |
| CN112365023A (en) * | 2020-09-30 | 2021-02-12 | 浙江汉德瑞智能科技有限公司 | Airport group event prediction and early warning method based on machine learning |
| CN112862172A (en) * | 2021-01-29 | 2021-05-28 | 国网河南省电力公司漯河供电公司 | State network 95598 power failure complaint prediction method and device, computer equipment and storage medium |
| CN113205197A (en) * | 2021-04-29 | 2021-08-03 | 国网河南省电力公司漯河供电公司 | Vehicle warehousing reservation management system based on logistics data |
| CN113495909A (en) * | 2020-04-06 | 2021-10-12 | 中信百信银行股份有限公司 | Customer complaint single quantity early warning method and device, electronic equipment and storage medium |
| CN116401601A (en) * | 2023-04-14 | 2023-07-07 | 国网浙江省电力有限公司 | Power failure sensitive user preferential treatment method based on logistic regression model |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106980929A (en) * | 2017-01-05 | 2017-07-25 | 国网福建省电力有限公司 | A kind of power failure complaint risk Forecasting Methodology based on random forest |
| CN107220732A (en) * | 2017-05-31 | 2017-09-29 | 福州大学 | A kind of power failure complaint risk Forecasting Methodology based on gradient boosted tree |
| CN108182453A (en) * | 2018-01-17 | 2018-06-19 | 贵州电网有限责任公司 | A kind of user's power failure susceptibility sorting technique |
| CN109447364A (en) * | 2018-11-08 | 2019-03-08 | 国网湖南省电力有限公司 | Power customer based on label complains prediction technique |
| CN109858669A (en) * | 2018-12-25 | 2019-06-07 | 国网安徽省电力有限公司 | A kind of arrearage based on Information Entropy stops telegram in reply and complains tendency prediction technique |
-
2019
- 2019-08-07 CN CN201910723742.XA patent/CN110503249A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106980929A (en) * | 2017-01-05 | 2017-07-25 | 国网福建省电力有限公司 | A kind of power failure complaint risk Forecasting Methodology based on random forest |
| CN107220732A (en) * | 2017-05-31 | 2017-09-29 | 福州大学 | A kind of power failure complaint risk Forecasting Methodology based on gradient boosted tree |
| CN108182453A (en) * | 2018-01-17 | 2018-06-19 | 贵州电网有限责任公司 | A kind of user's power failure susceptibility sorting technique |
| CN109447364A (en) * | 2018-11-08 | 2019-03-08 | 国网湖南省电力有限公司 | Power customer based on label complains prediction technique |
| CN109858669A (en) * | 2018-12-25 | 2019-06-07 | 国网安徽省电力有限公司 | A kind of arrearage based on Information Entropy stops telegram in reply and complains tendency prediction technique |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113495909A (en) * | 2020-04-06 | 2021-10-12 | 中信百信银行股份有限公司 | Customer complaint single quantity early warning method and device, electronic equipment and storage medium |
| CN111553816A (en) * | 2020-04-20 | 2020-08-18 | 北京北大软件工程股份有限公司 | Method and device for analyzing administrative review influence factors |
| CN111553816B (en) * | 2020-04-20 | 2023-11-03 | 北京北大软件工程股份有限公司 | Administrative multiple-proposal influence factor analysis method and device |
| CN111353792A (en) * | 2020-05-25 | 2020-06-30 | 广东电网有限责任公司惠州供电局 | Client portrait system with visual display and data analysis functions |
| CN112365023A (en) * | 2020-09-30 | 2021-02-12 | 浙江汉德瑞智能科技有限公司 | Airport group event prediction and early warning method based on machine learning |
| CN112862172A (en) * | 2021-01-29 | 2021-05-28 | 国网河南省电力公司漯河供电公司 | State network 95598 power failure complaint prediction method and device, computer equipment and storage medium |
| CN112862172B (en) * | 2021-01-29 | 2024-05-28 | 国网河南省电力公司漯河供电公司 | National network 95598 power outage complaint prediction method, device, computer equipment and storage medium |
| CN113205197A (en) * | 2021-04-29 | 2021-08-03 | 国网河南省电力公司漯河供电公司 | Vehicle warehousing reservation management system based on logistics data |
| CN113205197B (en) * | 2021-04-29 | 2023-04-07 | 国网河南省电力公司漯河供电公司 | Vehicle warehousing reservation management system based on logistics data |
| CN116401601A (en) * | 2023-04-14 | 2023-07-07 | 国网浙江省电力有限公司 | Power failure sensitive user preferential treatment method based on logistic regression model |
| CN116401601B (en) * | 2023-04-14 | 2023-09-15 | 国网浙江省电力有限公司 | Power failure sensitive user handling method based on logistic regression model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110503249A (en) | One kind complaining prediction technique caused by having a power failure | |
| Thom et al. | Spatiotemporal anomaly detection through visual analysis of geolocated twitter messages | |
| US10504194B2 (en) | Correlating consumption and voltage data to identify line loss in an electric grid | |
| CN106327055B (en) | A kind of electricity expense control method and system based on big data technology | |
| CN110334111B (en) | Multidimensional track analysis method and device | |
| Montero et al. | Fusing mobile phone data with other data sources to generate input OD matrices for transport models | |
| CN106093707A (en) | Intelligent electricity anti-theft analyzes the data processing method of system | |
| CN110210981A (en) | Insurance business Claims Resolution method, apparatus, computer installation and storage medium | |
| CN112860769B (en) | Energy planning data management system | |
| CN104484600B (en) | Intrusion detection method and device based on improved density clustering | |
| CN109063066A (en) | Data query method and device, data management system | |
| CN109934268A (en) | Abnormal transaction detection method and system | |
| CN109039710A (en) | Route data auditing method, device, server and storage medium | |
| CN103778041B (en) | Data monitoring method and device for internet | |
| CN108460698A (en) | Build the method and computer readable storage medium of patent value degree evaluation model | |
| CN110322688A (en) | A kind of method of data processing, the method for data query and relevant device | |
| CN110570097A (en) | business personnel risk identification method and device based on big data and storage medium | |
| CN108694462A (en) | Patent value degree evaluation method and computer readable storage medium | |
| CN108053095A (en) | A kind of electrical energy power quality disturbance affair character extracting method and system | |
| CN112330227A (en) | Urban management capability evaluation method and system based on digital urban management service construction | |
| CN117093617A (en) | Rail transit data analysis method, system, storage medium and electronic equipment | |
| CN110555452A (en) | network problem processing method and device based on intelligent clustering | |
| CN115358522A (en) | Enterprise online monitoring system and method | |
| CN105069665A (en) | Sorting and display method and apparatus of multi-channel off-site advertisement report data | |
| Domashova et al. | Detection of non-typical users of the electronic marketplace" Freight transportation" to prevent the competitive intelligence |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191126 |