CN107635004A - A kind of personalized service method for customizing in intelligent domestic system - Google Patents
A kind of personalized service method for customizing in intelligent domestic system Download PDFInfo
- Publication number
- CN107635004A CN107635004A CN201710879788.1A CN201710879788A CN107635004A CN 107635004 A CN107635004 A CN 107635004A CN 201710879788 A CN201710879788 A CN 201710879788A CN 107635004 A CN107635004 A CN 107635004A
- Authority
- CN
- China
- Prior art keywords
- user
- service
- forecast model
- data set
- intelligent domestic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses the personalized service method for customizing in a kind of intelligent domestic system, by analyzing servicing the historical data used in intelligent domestic system, predict interest-degree of the user to each service, so as to which customization of providing personalized service is to reduce the interaction time of intelligent domestic system and user, the function and performance of respective services in existing intelligent domestic system are not interfered with.The prediction to user interest degree is realized using the regression analysis of increment type, can ensure that when data volume increase or data are assembled new user interest degree forecast model can be calculated according to existing forecast model, both time and the space cost of forecast model calculating had been reduced, it is also ensured that the validity of new forecast model.The inventive method use range is extensive, the personalized service customization that both can be used in intelligent domestic system, can also be applied to the personalized function customization of special services in intelligent domestic system.
Description
Technical field
The invention belongs to the personalized service customization side in smart home field, more particularly to a kind of intelligent domestic system
Method.
Background technology
With the popularization of development and the application of smart home technology, user is increasingly dependent on intelligent domestic system, simultaneously
The user data that intelligent domestic system is collected into is more and more.Improve the approach of smart home Consumer's Experience gradually by traditional work(
Energy property increase turns to personalized service customization, i.e., the smart home service of personalization is provided according to the habits and customs of user, and
Ignore the uninterested service of user, so as to optimize usage experience of the user to intelligent domestic system.
Existing intelligent domestic system is gradually from tailored version to universal development, by electrical equipment control, environmental monitoring, data point
The various functions such as analysis are incorporated into a system, there is provided a large amount of general smart home services, the species of these services is increasingly
More, control is more and more finer, and operation becomes increasingly complex.But for many users, generally only to partial service therein
It is interested, and for some services lose interest in possibly even completely without using.Existing intelligent domestic system is supplied to each
The service of user is all available services, and user needs to take much time, and to grope which service be that oneself is interested
, and slowly these services, drop are found in search in a large amount of services for being required to when servicing using these every time provide from system
The low Consumer's Experience of intelligent domestic system.
Personalized service customization needs to predict it to intelligent domestic system according to user profile in intelligent domestic system
There is provided various services interest-degree, the priority application higher to user interest degree recommend, it is user-friendly, and for
Interest-degree relatively low application in family is hidden, and only user actively proposes explicitly to provide the service during use demand again.It is personalized
Service customization can accelerate interaction between user and intelligent domestic system, improve the Consumer's Experience of intelligent domestic system.It is existing
The mode for improving Consumer's Experience in some intelligent domestic systems mainly has three kinds：First, to various special smart home services
Integrate, a variety of intelligent controls are realized with a system, it is such as that the control of air-conditioning, refrigerator, electric cooker, curtain and electric light equipment is whole
Close into an intelligent domestic system；Second, predefined scene mode realizes the one-touch control to domestic environment, as user can
To pre-set the various kinds of equipment state for needing to control in the different scene such as the pattern of receiving a visitor, entertainment mode, and each scene,
It need to only select scene that the control to plurality of devices can be achieved during use；Third, the efficiency and quality of all kinds of services of optimization, mainly
Optimal Control Strategy, data processing algorithm, user interface etc., such as design more efficient algorithm realize to air conditioner refrigeration effect, when
Between equilibrium between power consumption.
The methods of existing all kinds of raising intelligent domestic system Consumer's Experiences although provide more fully systemic-function and
More efficient service quality, but also result in that all kinds of services in intelligent domestic system are more and more, and user is difficult in the short time
Interior which service of discovery is that oneself is interested, and also is difficult to that in each use oneself clothes interested can be quickly found out
Business.
The existing method for improving Consumer's Experience mainly includes function integration in intelligent domestic system, control is integrated and service
Optimize three classes.Various smart home service assemblies into same system, are easy to system pipes by function integration class method
Reason and user use, but cause user to be difficult quickly to determine which service is oneself because intelligent domestic system service is too many
It is interested, it is necessary to which multiple trial can just find the higher service of user interest degree.Moreover, using intelligent domestic system every time
When, even if user determines service interested, it is also difficult to rapidly find the service of oneself needs in a large amount of services.Control
Class and service optimization class method are integrated mainly with the execution efficiency for being used for the service of improving, the personalized service customization with user does not have
There is direct relation.
Most important step is prediction user to every during service customization personalized in realizing intelligent domestic system
The interest-degree of individual service.The prediction of user interest degree studied in intelligent domestic system it is more, it is each in generally use machine learning
The classical Forecasting Methodology of class, forecast model is obtained by the training to a large amount of historical datas, is carried out further according to forecast model specific
The interest-degree prediction of user.But in intelligent domestic system existing user interest degree Forecasting Methodology in training dataset increment more
Need to obtain new forecast model to the training dataset progress re -training after renewal under news.But in smart home
The scale of training dataset continuous updating and data set is increasing in system, existing Forecasting Methodology recalculate every time it is pre-
Need to expend substantial amounts of room and time cost when surveying model, especially when training dataset is on a grand scale, existing prediction side
Method can not quickly obtain new forecast model.
The content of the invention
Present invention mainly solves the service customization that personalization how is carried out in intelligent domestic system, reduces user and intelligence
The interaction time of house system, improve Consumer's Experience.Specifically, in terms of the problem to be solved in the present invention includes following three：
First, how by predicting user to the interest-degree of smart home service to improve the Consumer's Experience of intelligent domestic system；Second, as
The higher user interest degree forecast model of precision is quickly obtained in the historical data what is used from smart home service；Third, when instruction
When practicing collection incremental update occurs, incremental data and original forecast model how are only analyzed and the new user interest of quick obtaining
Forecast model is spent, without being recalculated to all training datasets.
The purpose of the present invention is achieved through the following technical solutions：Individual character in intelligent domestic system of the present invention
It is mainly individual to realize to the interest-degree of many services in intelligent domestic system by predicting each user to change service customization method
Property service customization, i.e. keypoint recommendation user service interested hides user uninterested service, avoids user each
It is required to select service interested from substantial amounts of service during using intelligent domestic system, reduces user and intelligent domestic system
Interaction time, so as to improve Consumer's Experience.
The overall procedure of personalized service method for customizing is broadly divided into three steps in intelligent domestic system of the present invention
Suddenly：Calculate the user interest degree forecast model of each service using data according to the history of respective services first, further according to work as
Preceding user profile calculates interest-degree of the user to all services；Then these services are arranged by the interest-degree descending of the user
Row；Finally system is optimized, the service higher to interest-degree ensures that user uses intelligent family every time to user's keypoint recommendation
Its service interested can be quickly found out when occupying system, and processing is then hidden for the relatively low service of interest-degree, when
The service for hiding Options and finding and needing to use is opened when user needs to use again.
User preference in intelligent domestic system of the present invention in personalized service method for customizing is predicted, it is necessary to predict
Interest-degree of the user to each single item service in intelligent domestic system.Predict that user includes following step to the interest-degree of a certain service
Suddenly：
Step 1：The foundation of forecast model.Assuming that the service condition data set D of the service records comprising N bars, per data
Represent a user useriTo the service condition of the service, specific form is (yi,Xi), wherein Xi=(xi1,xi2,...,xit)
Represent with whether using the related user property of the service, such as age, income, yiUser user is represented for 0-1 variables, 1i
Serviced using this, 0 represents user useriServiced without using this.Data set D is divided using the method for logistic regression
Analysis, regression coefficient θ=(θ0,θ1,…,θt).For given data set D, using the Newton iteration method of numerical analysis
(Newton-Raphson method) solves equation groupIt can obtain optimal recurrence system
Number
Step 2：According to active user user personal information Xu=(xu1,xu2,...,xut), and optimal recurrence system
NumberThe probability that user user uses the service is calculatedAs interest of the user user to the service
Degree；
Step 3：Solve the parameter of data set D forecast modelEach data set D's
Forecast model includes regression coefficientWith parameter F (D),For predicting interest-degree of the user to service, F (D) is used for increment more
The regression coefficient of new forecast model is calculated when new.
Step 4：New forecast model is calculated after data set D incremental updates.
Incremental update of the present invention to data set D includes two classes：First, increment class updates, i.e. D '=D ∪ Δ D,
Wherein data set D forecast model, it is known that and incremental data set Δ D and renewal after data set D' forecast model it is unknown,
User data set D such as Monday to Saturday has analyzed to obtain forecast model, and the data Δ D on Sunday has just updated, how to this
The data D' of all (Monday to Sunday) establishes forecast model；Second, aggregation class renewal, i.e. D '=D1∪D2∪…∪DK, wherein counting
According to collection D1、D2、DKDeng forecast model, it is known that and update after D' forecast model it is unknown, the aggregation class in intelligent domestic system
Renewal includes the data aggregation of time dimension and Spatial Dimension, as the data set in K city has established prediction in one week in the past
How model, obtain over the forecast models of data of whole region (including K city) in one week.
For aggregation class renewal D '=D1∪D2∪…∪DKIf data set D1、D2、...、DKForecast model be respectively
{R1,R2,...,RK, and the forecast model of the data set D' after assembling is Ra.Wherein R1、R2、...、RK, it is known that and RaIt is unknown.Ask
Solve RaThe step of it is as follows：
4.1：For given data collection DkForecast model RkFor, regression coefficient has been obtained in step 1 and step 3
And parameterI.e.To solve journey groupSolution,
Wherein N is data set DkTuple quantity, 1≤k≤K, K>1,1≤i≤N, tuple format are (yki,Xki), ykiRepresent data set
DkMiddle user useriWhether the service, X are usedki=(xki1,xki2,...,xkit) represent user useriAttribute in whether making
With the related attribute set of the service.
4.2：The forecast model for solving the data set D' after aggregation is RaParameterSpecific solution procedure
For：
In the first increment class updates D '=D ∪ Δs D, the recurrence system of renewal amount can be first solved according to step 1 and step 3
NumberAnd then data set D' forecast model R is calculatedaParameter Wherein F (a)=F
(D)+F (Δ),
The beneficial effects of the invention are as follows：
1. the method for the invention, by analyzing servicing the historical data used in intelligent domestic system, prediction is used
Family is to the interest-degree of each service, so as to provide personalized service customization to reduce intelligent domestic system with user when interacting
Between, do not interfere with the function and performance of respective services in existing intelligent domestic system.
, can be with 2. realize prediction to user interest degree using the regression analysis of increment type in the method for the invention
Ensure that new user interest degree prediction can be calculated according to existing forecast model when data volume increase or data are assembled
Model, time and the space cost of forecast model calculating were both reduced, it is also ensured that the validity of new forecast model.
3. the use range of the method for the invention is extensive, the personalized service that both can apply in intelligent domestic system
Customization, the personalized function customization of special services in intelligent domestic system can also be applied to.
Brief description of the drawings
Fig. 1 is the overview flow chart of personalized service method for customizing in intelligent domestic system；
Fig. 2 is that when time dimension and the enterprising row data of Spatial Dimension are assembled, user interest degree is predicted in intelligent domestic system
The incremental update schematic diagram of model.
Embodiment
The present invention is described in further detail with specific embodiment below in conjunction with the accompanying drawings.
Fig. 1 is the overall procedure of personalized service method for customizing in intelligent domestic system, to each in intelligent domestic system
The user of service establishes forecast model of the user to the interest-degree of each service using data set using the method for logistic regression, when
User uses data set incremental update or the renewal for be predicted again during data aggregation model.For current intelligent family
For occupying system user, according to user profile and the user interest degree forecast model of current each service, user is to every for prediction
The interest-degree of individual user, then these applications are finally subjected to intelligent domestic system according to the descending sort of the interest-degree predicted
Personalized service customization, i.e. the higher service of keypoint recommendation interest-degree.Prediction user specifically wraps to the interest-degree of a certain service
Containing following steps：
Step 1：The foundation of forecast model.Assuming that the service condition data set D of the service records comprising N bars, per data
Represent a user useriTo the service condition of the service, specific form is (yi,Xi), wherein Xi=(xi1,xi2,...,xit)
Represent with whether using the related user property of the service, such as age, income, yiUser user is represented for 0-1 variables, 1i
Serviced using this, 0 represents user useriServiced without using this.Data set D is divided using the method for logistic regression
Analysis, regression coefficient θ=(θ0,θ1,…,θt).For given data set D, using the Newton iteration method of numerical analysis
(Newton-Raphson method) solves equation groupIt can obtain optimal recurrence system
Number
Step 2：According to active user user personal information Xu=(xu1,xu2,...,xut), and optimal recurrence system
NumberThe probability that user user uses the service is calculatedIt is designated as interest of the user user to the service
Degree.
Step 3：Solve the parameter of data set D forecast modelEach data set D's
Forecast model includes regression coefficientWith parameter F (D),For predicting interest-degree of the user to service, F (D) is used for increment more
The regression coefficient of new forecast model is calculated when new.
Step 4：New forecast model is calculated after data set D incremental updates.
Incremental update of the present invention to data set D includes two classes：First, increment class updates, i.e. D '=D ∪ Δ D,
Wherein data set D forecast model, it is known that and incremental data set Δ D and renewal after data set D' forecast model it is unknown,
User data set D such as Monday to Saturday has analyzed to obtain forecast model, and the data Δ D on Sunday has just updated, how to this
The data D' of all (Monday to Sunday) establishes forecast model；Second, aggregation class renewal, i.e. D '=D1∪D2∪…∪DK, wherein counting
According to collection D1、D2、DKDeng forecast model, it is known that and update after D' forecast model it is unknown, the aggregation class in intelligent domestic system
Renewal includes the data aggregation of time dimension and Spatial Dimension, as the data set in K city has established prediction in one week in the past
How model, obtain over the forecast models of data of whole region (including K city) in one week.
For aggregation class renewal D '=D1∪D2∪…∪DKIf data set D1、D2、...、DKForecast model be respectively
{R1,R2,...,RK, and the forecast model of the data set D' after assembling is Ra.Wherein R1、R2、...、RK, it is known that and RaIt is unknown.Ask
Solve RaThe step of it is as follows：
4.1：For given data collection DkForecast model RkFor, regression coefficient has been obtained in step 1 and step 3
And parameterI.e.To solve journey groupSolution,
Wherein N is data set DkTuple quantity, 1≤k≤K, K>1,1≤i≤N, tuple format are (yki,Xki), ykiRepresent data set
DkMiddle user useriWhether the service, X are usedki=(xki1,xki2,...,xkit) represent user useriAttribute in whether making
With the related attribute set of the service.
4.2：The forecast model for solving the data set D' after aggregation is RaParameterSpecific solution procedure
For：
In the first increment class updates D '=D ∪ Δs D, the recurrence system of renewal amount can be first solved according to step 1 and step 3
NumberAnd then data set D' forecast model R is calculatedaParameter Wherein F (a)=F
(D)+F (Δ),
Fig. 2 is that when time dimension and the enterprising row data of Spatial Dimension are assembled, user interest degree is predicted in intelligent domestic system
The incremental update of model.The incremental update of interest-degree forecast model on time dimension as shown in dash box on the left of Fig. 2, for
For some region, such as region M1 in area 1, if time T1...TkEvery day is represented respectively, it is caused daily according to the region
The forecast model that smart home service service condition is calculated is respectively { R11,R12,...,R1K}.To the region one week
Interior smart home service service condition is analyzed, i.e. K=7, then needs the forecast model { R known to11,R12,...,
R1K, the forecast model R after renewal is calculatedM1.To enter to the smart home service service condition in whole distract 1 one weeks
Row analysis, then need according to forecast model { RM1,RM2,...,RMNNew forecast model R ' is further calculatedA1。
Electricity consumption behavioural analysis on Spatial Dimension is as shown in dash box above Fig. 2, to time T1For interior area 1,
Smart home service service condition of each region in time T1 is analyzed to obtain the interest-degree forecast model in N number of region
{R11,R21,...,RN1, after the incremental update calculating of model is predicted to it, it can obtain whole after progress space clustering
The regional 1 intelligent domestic system service in time T1 uses the user interest degree forecast model R of dataA1.To all
The intelligent domestic system service service condition in area is analyzed, then is needed further according to Q area in T1The user interest degree at moment
Forecast model { RA1,RA2,...,RAQCalculated, all regions are obtained in T1The forecast model R at momentB1。
The inventive method can independently carry out the service customization of personalization with user profile, accelerate user and intelligent domestic system
Interactive speed, improve the Consumer's Experience of intelligent domestic system.User interest degree forecast model can in training set incremental update
To be directly predicted the incrementally updating of model, without being predicted model to whole data set in each incremental update
Recalculate, both ensure that the validity of forecast model also improves the renewal efficiency of forecast model.The inventive method is being instructed
Practice the renewal that collection data can directly be predicted model after Spatial Dimension and time dimension carry out data aggregation, without carrying out
Recalculating for model is predicted again after the data aggregation of training set.The inventive method is applicable not only to intelligent domestic system
The individualized selection of middle respective services, it can be used for the individualized selection of respective services function in single smart home service.
Claims (1)
1. the personalized service method for customizing in a kind of intelligent domestic system, it is characterised in that comprise the following steps：
Step 1：The user interest degree of each service is calculated using data according to the history of respective services in intelligent domestic system
Forecast model：Assuming that the service condition data set D of the service records comprising N bars, a user user is represented per dataiTo this
The service condition of service, specific form are (yi,Xi), wherein Xi=(xi1,xi2,...,xit) represent with whether being had using the service
The user property of relation, yiUser user is represented for 0-1 variables, 1iServiced using this, 0 represents user useriWithout using this
Service；Data set D is analyzed using the method for logistic regression, regression coefficient θ=(θ0,θ1,…,θt)；Using numerical analysis
Newton iteration method solve equation groupIt can obtain optimal regression coefficient
Step 2：According to active user user personal information Xu=(xu1,xu2,...,xut), and optimal regression coefficient
The probability that user user uses the service is calculatedAs interest-degrees of the user user to the service；
Step 3：Solve the parameter of data set D forecast modelEach data set D prediction mould
Type includes regression coefficientWith parameter F (D),For predicting interest-degree of the user to service, F (D) is used to calculate during incremental update
The regression coefficient of new forecast model；
Step 4：New forecast model is calculated after data set D incremental updates, incremental update includes two classes：First, increment class updates,
That is D '=D ∪ Δ D, wherein data set D forecast model, it is known that and incremental data set Δ D and renewal after data set D' it is pre-
It is unknown to survey model；Second, aggregation class renewal, i.e. D '=D1∪D2∪…∪DK, wherein data set D1、D2、…、DKPrediction mould
Type is, it is known that and D' forecast model is unknown after updating；
For aggregation class renewal D '=D1∪D2∪…∪DKIf data set D1、D2、...、DKForecast model be respectively { R1,
R2,...,RK, and the forecast model of the data set D' after assembling is Ra；Solve RaThe step of it is as follows：
4.1：For given data collection DkForecast model Rk, regression coefficient and parameter have been obtained in step 1 and step 3I.e.To solve journey groupSolution,
Wherein N is data set DkTuple quantity, 1≤k≤K, K>1,1≤i≤N, tuple format are (yki,Xki), ykiRepresent data set
DkMiddle user useriWhether the service, X are usedki=(xki1,xki2,...,xkit) represent user useriAttribute in whether making
With the related attribute set of the service；
4.2：The forecast model for solving the data set D' after aggregation is RaParameterSpecifically solution procedure is：
D '=D ∪ Δ D are updated for increment class, the regression coefficient of renewal amount is first solved according to step 1 and step 3
Then data set D' forecast model R is calculatedaParameterWherein F (a)=F (D)+F (Δ),
Step 5：The interest-degree of all services is optimized to system according to user user, to user's keypoint recommendation interest-degree compared with
High service, the relatively low service of interest-degree is hidden, realize the personalized service customization of intelligent domestic system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710879788.1A CN107635004B (en) | 2017-09-26 | 2017-09-26 | Personalized service customization method in intelligent home system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710879788.1A CN107635004B (en) | 2017-09-26 | 2017-09-26 | Personalized service customization method in intelligent home system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107635004A true CN107635004A (en) | 2018-01-26 |
| CN107635004B CN107635004B (en) | 2020-12-08 |
Family
ID=61102362
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710879788.1A Active CN107635004B (en) | 2017-09-26 | 2017-09-26 | Personalized service customization method in intelligent home system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107635004B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109212988A (en) * | 2018-09-21 | 2019-01-15 | 中国联合网络通信集团有限公司 | Intelligent home furnishing control method and system |
| CN111563203A (en) * | 2020-05-08 | 2020-08-21 | 深圳市万佳安人工智能数据技术有限公司 | Intelligent household user-service interest degree personalized prediction device and method based on rapid non-negative implicit characteristic analysis |
| CN111669304A (en) * | 2020-05-19 | 2020-09-15 | 广东科徕尼智能科技有限公司 | Intelligent household scene control method and equipment based on edge gateway and storage medium |
| CN114326396A (en) * | 2021-12-24 | 2022-04-12 | 珠海格力电器股份有限公司 | Model adjusting method and device and electronic equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005059791A1 (en) * | 2003-12-15 | 2005-06-30 | Koninklijke Philips Electronics N.V. | Communication method and system using priority technology |
| CN1967533A (en) * | 2006-07-17 | 2007-05-23 | 北京航空航天大学 | Gateway personalized recommendation service method and system introduced yuan recommendation engine |
| CN103488705A (en) * | 2013-09-06 | 2014-01-01 | 电子科技大学 | User interest model incremental update method of personalized recommendation system |
| CN104834967A (en) * | 2015-04-24 | 2015-08-12 | 南京邮电大学 | User similarity-based business behavior prediction method under ubiquitous network |
-
2017
- 2017-09-26 CN CN201710879788.1A patent/CN107635004B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005059791A1 (en) * | 2003-12-15 | 2005-06-30 | Koninklijke Philips Electronics N.V. | Communication method and system using priority technology |
| CN1967533A (en) * | 2006-07-17 | 2007-05-23 | 北京航空航天大学 | Gateway personalized recommendation service method and system introduced yuan recommendation engine |
| CN103488705A (en) * | 2013-09-06 | 2014-01-01 | 电子科技大学 | User interest model incremental update method of personalized recommendation system |
| CN104834967A (en) * | 2015-04-24 | 2015-08-12 | 南京邮电大学 | User similarity-based business behavior prediction method under ubiquitous network |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109212988A (en) * | 2018-09-21 | 2019-01-15 | 中国联合网络通信集团有限公司 | Intelligent home furnishing control method and system |
| CN109212988B (en) * | 2018-09-21 | 2021-07-06 | 中国联合网络通信集团有限公司 | Intelligent household control method and system |
| CN111563203A (en) * | 2020-05-08 | 2020-08-21 | 深圳市万佳安人工智能数据技术有限公司 | Intelligent household user-service interest degree personalized prediction device and method based on rapid non-negative implicit characteristic analysis |
| CN111669304A (en) * | 2020-05-19 | 2020-09-15 | 广东科徕尼智能科技有限公司 | Intelligent household scene control method and equipment based on edge gateway and storage medium |
| CN114326396A (en) * | 2021-12-24 | 2022-04-12 | 珠海格力电器股份有限公司 | Model adjusting method and device and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107635004B (en) | 2020-12-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107635004A (en) | A kind of personalized service method for customizing in intelligent domestic system | |
| Himeur et al. | A survey of recommender systems for energy efficiency in buildings: Principles, challenges and prospects | |
| CN107120782B (en) | A kind of HVAC system control method based on multi-user's hot comfort data | |
| CN100412870C (en) | Gateway personalized recommendation service method and system introduced yuan recommendation engine | |
| CN103106230A (en) | Information processing apparatus, information processing method, and program | |
| Ponce et al. | Framework for communicating with consumers using an expectation interface in smart thermostats | |
| CN112465385B (en) | Demand response potential analysis method applying intelligent ammeter data | |
| Alsalemi et al. | Smart energy usage and visualization based on micro-moments | |
| Zhou et al. | A Comprehensive Review of the Applications of Machine Learning for HVAC | |
| Kannan et al. | Artificial intelligence based air conditioner energy saving using a novel preference map | |
| CN108694211B (en) | Application distribution method and device | |
| Li | System simulation of driving mechanism of rural tourism development based on data mining analysis and cloud computing | |
| Jrhilifa et al. | Multi-horizon short-term load consumption forecasting using deep learning models | |
| Michelson et al. | The theoretical status and operational usage of life style in environmental research | |
| KR20200094834A (en) | Platform for gathering information for ai entities and method by using the same | |
| Liao et al. | Location prediction through activity purpose: integrating temporal and sequential models | |
| Tao et al. | Discrete imperialist competitive algorithm for the resource-constrained hybrid flowshop problem | |
| Wang et al. | Research on energy consumption in household sector: a comprehensive review based on bibliometric analysis | |
| CN112255923A (en) | Electric equipment control method, device, server and medium | |
| He et al. | Review of swarm intelligence algorithms for multi-objective flowshop scheduling | |
| Sitahong et al. | Learning dispatching rules via novel genetic programming with feature selection in energy-aware dynamic job-shop scheduling | |
| Serafini et al. | Optimized development: defining design rules through product optimization techniques | |
| Shalaby et al. | A Prototype Model of Monitoring Energy Consumption and Optimizing Distribution of Smart Buildings | |
| Han et al. | Predicting energy use in construction using Extreme Gradient Boosting | |
| Vaz et al. | Merging entropy in self-organisation: a geographical approach |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20201016 Address after: 310000 room 2201, Huafeng international business building, Jianggan District, Hangzhou City, Zhejiang Province Applicant after: HANGZHOU KONKE INFORMATION TECHNOLOGY Co.,Ltd. Address before: 322000, No. 968 Xuefeng West Road, Jinhua, Zhejiang, Yiwu Applicant before: YIWU KONKE TECHNOLOGY Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |