CN113158027A - Intelligent device recommendation method and system and intelligent terminal - Google Patents
Intelligent device recommendation method and system and intelligent terminal Download PDFInfo
- Publication number
- CN113158027A CN113158027A CN202110255243.XA CN202110255243A CN113158027A CN 113158027 A CN113158027 A CN 113158027A CN 202110255243 A CN202110255243 A CN 202110255243A CN 113158027 A CN113158027 A CN 113158027A
- Authority
- CN
- China
- Prior art keywords
- information
- model
- recommendation
- user
- intelligent
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
An intelligent device recommendation method, a system and an intelligent terminal are provided, the method comprises the following steps: receiving information of multiple dimensions, wherein the information comprises equipment information, external environment information, internal environment information and terminal information; calculating the sequencing recommendation of a plurality of recommendation devices and the sequencing recommendation of a plurality of common operations of each recommendation device according to the information; displaying the calculation result and providing the calculation result for a user to select; and recording the selection result of the user. The invention forms a set of individual and accurate recommendation model of the intelligent equipment for the user, and can enable the user to quickly and accurately find the intelligent equipment to be operated at the intelligent terminal.
Description
Technical Field
The invention relates to the technical field of intelligent terminals, in particular to an intelligent device recommendation method and system and an intelligent terminal.
Background
In the era of smart phones, the sizes of mobile phones and smart home devices are rapidly growing in industrial scenes. In a future 5G large-scale Internet of things equipment erection scene, a user needs to spend a great deal of time on sliding an interface in an intelligent terminal to search for a single intelligent equipment, or to remember the interface where the equipment is located and the position where the equipment is located in the interface.
The past sorting algorithm only considers the use frequency of the equipment and the local information of the intelligent terminal. Under such an algorithm, low frequency devices will be placed at the bottom.
Disclosure of Invention
The invention aims to solve the problems and provides an intelligent device recommendation method, an intelligent device recommendation system and an intelligent terminal, which are used for sequencing and recommending intelligent devices by integrating information of multiple dimensions, forming a set of individual and accurate intelligent device recommendation model for a user and enabling the user to quickly and accurately find the intelligent device to be operated at the intelligent terminal.
In order to achieve the purpose, the invention adopts the technical scheme that:
an intelligent device recommendation method comprises the following steps: receiving information of multiple dimensions, wherein the information comprises equipment information, external environment information, internal environment information and terminal information; calculating the sequencing recommendation of a plurality of recommendation devices and the sequencing recommendation of a plurality of common operations of each recommendation device according to the information; displaying the calculation result and providing the calculation result for a user to select; and recording the selection result of the user.
Preferably, a machine learning algorithm is used for calculation, and the model of the machine learning algorithm is updated according to the selection result.
Preferably, each selection result of the user is recorded as a piece of data, when the data volume and the covered time interval of the data set respectively reach a threshold value, the data set is used for training the model to obtain a new model, and if the accuracy of the new model is higher than that of the original model, the new model is used for replacing the original model.
Preferably, the update strategy of the model of the machine learning algorithm adopts a viterbi algorithm.
Preferably, the accuracy of the new model is verified by the data of the previous training cycle or in real time.
Preferably, the model is updated during the intersection period of user inactivity periods and low computational load periods.
Preferably, the initial model of the model is a model based on a heuristic algorithm, or the initial model of the model is obtained by training.
Preferably, the machine learning algorithm is run in an intelligent terminal, a gateway or a cloud.
Based on the same inventive concept, the invention also provides an intelligent device recommendation system, which comprises: the information receiving module is used for receiving information of multiple dimensions, wherein the information comprises equipment information, external environment information, internal environment information and terminal information; the calculation module is used for calculating the sequencing recommendation of a plurality of recommendation devices and the sequencing recommendation of a plurality of common operations of each recommendation device according to the information; the display module is used for displaying the calculation result and providing the calculation result for the user to select; and the recording module is used for recording the selection result of the user.
Based on the same inventive concept, the invention also provides an intelligent terminal, wherein the intelligent terminal is used for running the program, and the intelligent device recommendation method is executed when the program runs.
The invention has the beneficial effects that:
(1) integrating information of multiple dimensions, and providing personalized and accurate sequencing recommendation of intelligent equipment for a user according to an intelligent scene;
(2) adopting a self-adaptive machine learning algorithm to obtain and record feedback in the process of continuous use of the user, wherein the algorithm learns the use habit of the user from the feedback periodically;
(3) the algorithm is operated in an intermediate node which is not limited to the intelligent terminal, and different algorithms can be selected according to the conditions of computational load of the intermediate node, the smoothness degree of a network and the like.
Drawings
FIG. 1 is a flow chart of a method for intelligent device recommendation;
FIG. 2 is a flow chart of a model update based on a collaborative filtering algorithm;
FIG. 3 is a flowchart of a model update based on a deep learning algorithm;
FIG. 4 is a flowchart of an update of a model based on the Viterbi Algorithm;
FIG. 5 is a flow chart of a model update period selection;
fig. 6 is a diagram of a ranking recommendation interface of an intelligent device of an intelligent terminal.
Detailed Description
In order to make the technical problems, technical solutions and advantageous effects to be solved by the present invention clearer and more obvious, the present invention is further described in detail with reference to specific embodiments below. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Example one
As shown in fig. 1, the present embodiment provides an intelligent device recommendation method, which can dynamically calculate and predict the intelligent devices that a specified user wants to operate, and sort and recommend the devices.
The method comprises the following steps:
receiving information of multiple dimensions, wherein the information comprises equipment information, external environment information, internal environment information and terminal information.
The method described in this embodiment is performed in a system, the system includes an intermediate node and one or more adaptive machine learning algorithms, the machine learning algorithm is performed in the intermediate node, and the intermediate node may be a cloud, a high-performance gateway, or an intelligent terminal, and is not limited to an intelligent terminal in a conventional algorithm, that is, is not limited to a local processing algorithm process.
The selection of the algorithm described in this embodiment depends on the network congestion of the system erection area, the cloud server load, the size of the sensor network, and other conditions, and the performance budget of the system.
If the computing power, storage space, or energy supply of the intermediate nodes is limited, then conventional machine learning algorithms, including but not limited to decision trees, Support Vector Machines (SVMs), small-scale neural network algorithms such as linear/logistic regression, bayesian statistics, or recommendation algorithms with fewer dimensions or indices may be used.
If the computational power storage, space or energy supply of the intermediate nodes is sufficient, deep learning algorithms, including but not limited to large scale neural network algorithms such as Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN) or long short term memory networks (LSTM), may be used, and various neural networks may be used in cascade if necessary, or recommendation algorithms with relatively more dimensions or indices may be used.
The algorithm described in this embodiment is a sort recommendation algorithm. The system described in this embodiment receives information in multiple dimensions as input to the algorithm. Such information may include, but is not limited to, the following four categories:
first type, device information: the method includes attributes, states, and mutual association relations of the intelligent devices in the intelligent scene (such as a home scene, a work scene, etc.), including but not limited to the location, type, associated device configuration, device activity information, etc. of the intelligent devices. Wherein the device activity information may refer to a particular interactive device that has performed a particular operation for a previous period of time.
Second type, external environment information: such as current weather, external geographical environment, specific service information, value added service information, emergency, geographical information, etc. Such information may be invested by external information sources, including weather service providers, meteorological departments, value added service related departments, emergency notification systems, and the like.
Third type, internal environment information: and information of the intelligent environment where the terminal is located. This information is often provided by internal sensors, and may include:
simple sensor information such as temperature, humidity, room brightness, etc.;
for example, the door opening and closing records of the intelligent door lock product, such as the number of the indoor population calculated, the indoor area calculated by the number of the specific intelligent devices, the population density calculated by the face recognition system, and the like.
Such as indoor space information drawn by intelligent sweeping and mopping robots.
Various environment, configuration information measured or entered manually.
Fourth, terminal information:
information of an account to which the intelligent terminal belongs, such as objective information of user identity, height, weight and the like and subjective information of preference and the like;
the position and the angle of the intelligent terminal are directly obtained or calculated information is gradually deduced through the sensor.
This information may be pre-determined and pre-processed to some extent before being subjected to algorithmic calculations.
And the intermediate node calculates one or a plurality of intelligent devices with high weight values through a sorting recommendation algorithm according to the information to serve as recommendation devices, namely dynamically calculating and predicting the devices to be operated next by the appointed user, and sorting and recommending the devices which are possible to be operated. Further, the sequencing recommendation of one or a plurality of common operations of each recommendation device is calculated.
The method provided by the embodiment integrates information of multiple dimensions as the input of the algorithm, and is not limited to the local information of the intelligent terminal in the traditional method, so that an intelligent scene for a user to operate the intelligent device is constructed, and personalized and accurate sequencing recommendation of the intelligent device is provided for the user.
And displaying the calculation result and providing the calculation result for the user to select. And based on the sorting and recommendation results, the user interface of the intelligent terminal displays one or a plurality of weighting devices in priority. The user interface may be an APP, web site, or other customized user interface.
And recording the selection result of the user. Each user's selection made on the intelligent terminal will be recorded as a piece of raw data of the personal model. The preprocessed data will be added to the training set of the individual. When the amount of data reaches a certain level and the data set covers a sufficiently long time interval, the training set will be used, in whole or in part, to train the personalized model.
The present embodiment employs a general machine learning process to build a model of the algorithm. For the initial state of the personal model for each user, an initial model with superior performance and resource consumption may be selected as the default.
In this embodiment, the model is obtained by a general process training, that is, a part of data (training set) is input for model training, and the performance of the trained model is verified by another part of data (evaluation set disjoint to the training set).
In other embodiments, the initial model may be a simpler model based on a default recommendation heuristic.
The system described in this embodiment obtains and records feedback during the continuous use of the user. The algorithm learns the use habits of the user from the feedback regularly, and finally forms a set of individual and accurate recommendation model for the user. The function interaction can enable a user to quickly and accurately find the equipment to be operated at the intelligent terminal, such as a mobile phone end, the selection difficulty of the user on the equipment is reduced at an application level, and the user experience is improved.
The model of the algorithm described in this embodiment can be continuously updated through the feedback of the user, and when the accuracy of the new model is higher than that of the original model, the new model is used to replace the original model as the active model.
Wherein the accuracy of the new model is verified using real-time (next cycle) or past (last update cycle as above) data as the evaluation set.
The model of the algorithm described in this embodiment adopts an update strategy of finite resource load balancing.
Those skilled in the art will understand that all or part of the steps in the above method embodiments may be implemented by a program to instruct related hardware, where the program is stored in a storage medium and includes several instructions to enable a device (which may be a single chip, a chip, etc.) or a processor (processor) to execute all or part of the steps of the method described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
Example two
As shown in fig. 2, the present embodiment provides an update strategy based on a model of a collaborative filtering algorithm, i.e., a conventional machine learning algorithm. In the initial phase, each user may be assigned an initial model that matches their base identity. After a certain use period (such as several weeks, months or quarterly), the data volume collected by the intermediate node and the time interval covered by the data set respectively reach the threshold values, retraining the model, adding the user data obtained in the updating period into the training set, and training to obtain a new model. And when the accuracy of the new model is higher than that of the original model, replacing the original model with the new model. The updating strategy has the advantages that the model is easy to train and deploy under the condition of low data volume and can be carried out locally at the intelligent terminal or at the middle node with limited performance, and the updating strategy has the defects that the model has certain performance upper limit, namely the marginal benefit of data investment is reduced, and when the training time is increased along with the increase of the data volume, the prediction precision of the model is improved only in a very limited way.
EXAMPLE III
As shown in fig. 3, the present embodiment provides an update strategy of a model based on a dynamic deep learning algorithm. In the initial phase, the system described in this embodiment uses a default recommendation algorithm model, such as LSTM or cascaded deep learning networks that have been pre-trained on other similar data sets. After a certain period of use (several weeks, months or quarterly), the intermediate node retrains the model by using the user data obtained in the updating period through a migration learning method, trains the parameters of the specific range of the model, obtains the migrated model, and provides personalized upgrade for the model. The updating strategy has the advantages that the model prediction precision is improved along with the increase of the data volume, and the defect is that the basic processing capacity of the intermediate node is required. Therefore, the algorithm under the updating strategy is more suitable for being operated in the cloud.
Example four
As shown in fig. 4, the present embodiment provides an update strategy based on a model of the viterbi algorithm. The present embodiment uses the viterbi algorithm to determine the time domain maximum likelihood path in reverse. Where the value of L, M may depend on the actual computational cost. And M also depends on the results of various preprocessing algorithms. In this embodiment, L is 3, and M.gtoreq.3. And when the stage number L is more than or equal to L, the intermediate node trains the parameters of the specific range of the model by using the user data obtained in the updating period, a new model is obtained by the migration of the current active model, the new model and the stored past models form a new model group, and the model with the maximum accuracy in the new model group is selected as the active model. And when the stage L is smaller than L, the intermediate node trains the parameters of the specific range of the model by using the user data obtained in the updating period to obtain a new model, and the model with the maximum accuracy rate in the new model and the stored past models is selected as an active model.
The updating strategy shown in the embodiment can obtain a model with more individuality and accuracy, and provides sequencing recommendation of the intelligent device for the user.
EXAMPLE five
The system in the first embodiment has a set of strategy for load balancing of comprehensive computing power and flow, and prevents massive training, computing and transmission requests from being bundled.
As shown in fig. 5, the present embodiment provides a selection strategy for a model update period of the system. Before the model updating is carried out, the intermediate node can run a program to judge the load of elements such as self computing power, flow and intelligent equipment using frequency of a user, and selects a proper time node for updating in an updating period.
The intermediate node queries for user inactivity periods and low computational load periods (which may not exist), respectively. If the two nodes have intersection, a new model training task is submitted to the intermediate node in the intersection period, and the model is updated. And if the intersection does not exist, selecting a starting node of the user inactivity time period to submit a new model training task.
EXAMPLE six
The embodiment provides an intelligent device recommendation system using the method in the first embodiment. The system comprises:
and the information receiving module is arranged at the intermediate node (which can be an intelligent terminal). The information receiving module is in communication connection with the intelligent terminal, the sensor network, the external information source and each intelligent device and receives information of multiple dimensions, wherein the information comprises device information, external environment information, internal environment information and terminal information.
And the computing module is arranged at the intermediate node and runs the machine learning algorithm in the first embodiment. The recommendation system is used for calculating the sequencing recommendations of a plurality of recommendation devices and the sequencing recommendations of a plurality of common operations of each recommendation device according to the information;
and the display module is arranged at the intelligent terminal. And the user interface of the display module displays the calculation result of the calculation module, preferentially displays one or more weighting devices and the common operation thereof, and provides the user with the selection.
And the recording module is arranged at the middle node. The recording module is used for recording the selection result of the user, each selection result is recorded as a piece of data, and when the data volume and the covered time interval of the data set reach threshold values respectively, the model is updated to obtain a new model.
EXAMPLE seven
The embodiment provides an intelligent terminal, which is used for running a program, wherein the method according to the first embodiment is executed when the program runs.
The intelligent terminal described in this embodiment may be a multifunctional intelligent terminal, and performs network or function interaction by opening an APP or calling up a UI, or may be a single-function terminal, and performs network or function interaction by using a fixed user interaction interface or a key. Input and output and display in the intelligent terminal can be bound with the user name and the subordinate intelligent equipment thereof.
As shown in fig. 6, the user interface of the intelligent terminal according to this embodiment may show the calculation result of the algorithm according to the first embodiment, and provide the user with ranking recommendation of the intelligent device and the common operations thereof.
While the above description shows and describes the preferred embodiments of the present invention, it is to be understood that the invention is not limited to the forms disclosed herein, but is not to be construed as excluding other embodiments and is capable of use in various other combinations, modifications, and environments and is capable of changes within the scope of the inventive concept as expressed herein, commensurate with the above teachings, or the skill or knowledge of the relevant art. And that modifications and variations may be effected by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. An intelligent device recommendation method is characterized by comprising the following steps:
receiving information of multiple dimensions, wherein the information comprises equipment information, external environment information, internal environment information and terminal information;
calculating the sequencing recommendation of a plurality of recommendation devices and the sequencing recommendation of a plurality of common operations of each recommendation device according to the information;
displaying the calculation result and providing the calculation result for a user to select;
and recording the selection result of the user.
2. The intelligent device recommendation method according to claim 1, wherein a machine learning algorithm is used for calculation, and a model of the machine learning algorithm is updated according to the selection result.
3. The intelligent device recommendation method according to claim 2, wherein each selection result of the user is recorded as a piece of data, when the data volume and the covered time interval of the data set respectively reach a threshold, the data set is used for training the model to obtain a new model, and if the accuracy of the new model is higher than that of the original model, the new model is used for replacing the original model.
4. The smart device recommendation method according to claim 3, wherein the update strategy of the model of the machine learning algorithm employs a Viterbi algorithm.
5. The smart device recommendation method according to claim 3, wherein the accuracy of the new model is verified by the data of the real-time or last training cycle.
6. The smart device recommendation method of claim 2, wherein the model is updated during an intersection period of user inactivity periods and computing load low periods.
7. The intelligent device recommendation method according to claim 2, wherein the initial model of the model is a model based on a heuristic algorithm, or the initial model of the model is obtained by training.
8. The intelligent device recommendation method according to claim 2, wherein the machine learning algorithm is executed in an intelligent terminal or a gateway or a cloud.
9. An intelligent device recommendation system, comprising:
the information receiving module is used for receiving information of multiple dimensions, wherein the information comprises equipment information, external environment information, internal environment information and terminal information;
the calculation module is used for calculating the sequencing recommendation of a plurality of recommendation devices and the sequencing recommendation of a plurality of common operations of each recommendation device according to the information;
the display module is used for displaying the calculation result and providing the calculation result for the user to select;
and the recording module is used for recording the selection result of the user.
10. An intelligent terminal, wherein the intelligent terminal is configured to run a program, and when the program runs, the intelligent device recommendation method according to any one of claims 1 to 8 is executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110255243.XA CN113158027A (en) | 2021-03-09 | 2021-03-09 | Intelligent device recommendation method and system and intelligent terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110255243.XA CN113158027A (en) | 2021-03-09 | 2021-03-09 | Intelligent device recommendation method and system and intelligent terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113158027A true CN113158027A (en) | 2021-07-23 |
Family
ID=76884442
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110255243.XA Pending CN113158027A (en) | 2021-03-09 | 2021-03-09 | Intelligent device recommendation method and system and intelligent terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113158027A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115952009A (en) * | 2023-03-15 | 2023-04-11 | 北京泰尔英福科技有限公司 | Data center recommendation method and device based on computational network fusion characteristics |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106682054A (en) * | 2016-05-24 | 2017-05-17 | 腾讯科技（深圳）有限公司 | Terminal application recommendation method, terminal application recommendation device and terminal application recommendation system |
| EP3435643A1 (en) * | 2017-07-29 | 2019-01-30 | Advanced Digital Broadcast S.A. | A system and method for control of an appliance by voice |
| CN110727494A (en) * | 2019-09-30 | 2020-01-24 | Oppo广东移动通信有限公司 | Application icon control method and related device |
| CN112051744A (en) * | 2019-08-15 | 2020-12-08 | 河南紫联物联网技术有限公司 | Intelligent terminal, and recommendation method and system for intelligent home application |
-
2021
- 2021-03-09 CN CN202110255243.XA patent/CN113158027A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106682054A (en) * | 2016-05-24 | 2017-05-17 | 腾讯科技（深圳）有限公司 | Terminal application recommendation method, terminal application recommendation device and terminal application recommendation system |
| EP3435643A1 (en) * | 2017-07-29 | 2019-01-30 | Advanced Digital Broadcast S.A. | A system and method for control of an appliance by voice |
| CN112051744A (en) * | 2019-08-15 | 2020-12-08 | 河南紫联物联网技术有限公司 | Intelligent terminal, and recommendation method and system for intelligent home application |
| CN110727494A (en) * | 2019-09-30 | 2020-01-24 | Oppo广东移动通信有限公司 | Application icon control method and related device |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115952009A (en) * | 2023-03-15 | 2023-04-11 | 北京泰尔英福科技有限公司 | Data center recommendation method and device based on computational network fusion characteristics |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020011068A1 (en) | Method and system for executing machine learning process | |
| Zhao et al. | Where to go next: A spatio-temporal LSTM model for next POI recommendation | |
| Angelov et al. | A new type of simplified fuzzy rule-based system | |
| Nguyen et al. | Multiple neural networks for a long term time series forecast | |
| CN108361927A (en) | A kind of air-conditioner control method, device and air conditioner based on machine learning | |
| EP2936334A1 (en) | Instance weighted learning machine learning model | |
| US11501215B2 (en) | Hierarchical clustered reinforcement machine learning | |
| JP2007317068A (en) | Recommending device and recommending system | |
| Han et al. | The reinforcement learning method for occupant behavior in building control: A review | |
| Assareh et al. | Forecasting energy demand in Iran using genetic algorithm (GA) and particle swarm optimization (PSO) methods | |
| CN109471370A (en) | A kind of behavior prediction and control method based on exhaust fan operation data, system | |
| CN116562514B (en) | Method and system for immediately analyzing production conditions of enterprises based on neural network | |
| CN111061959A (en) | Developer characteristic-based crowd-sourcing software task recommendation method | |
| CN111512299A (en) | Method for content search and electronic device thereof | |
| CN110781595A (en) | Energy use efficiency PUE prediction method, device, terminal and medium | |
| CN110008977A (en) | Clustering Model construction method and device | |
| Muccini et al. | Leveraging machine learning techniques for architecting self-adaptive iot systems | |
| CN113158027A (en) | Intelligent device recommendation method and system and intelligent terminal | |
| CN113688306A (en) | Recommendation strategy generation method and device based on reinforcement learning | |
| Rahimi-Vahed et al. | A hybrid multi-objective particle swarm algorithm for a mixed-model assembly line sequencing problem | |
| CN108563720A (en) | Big data based on AI recommends learning system and recommends method | |
| Khargharia et al. | Probabilistic analysis of context caching in Internet of Things applications | |
| Gong et al. | Interactive Genetic Algorithms with Individual Fitness Not Assigned by Human. | |
| Tegelund et al. | A task-oriented service personalization scheme for smart environments using reinforcement learning | |
| Aiolli et al. | Application of the preference learning model to a human resources selection task |
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: 20210723 |