CN110772797B - Data processing method, device, server and storage medium - Google Patents
Data processing method, device, server and storage medium Download PDFInfo
- Publication number
- CN110772797B CN110772797B CN201911037287.4A CN201911037287A CN110772797B CN 110772797 B CN110772797 B CN 110772797B CN 201911037287 A CN201911037287 A CN 201911037287A CN 110772797 B CN110772797 B CN 110772797B
- Authority
- CN
- China
- Prior art keywords
- user
- ability
- target
- target user
- capacity
- 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.)
- Active
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- A63F13/00—Video games, i.e. games using an electronically generated display having two or more dimensions
- A63F13/70—Game security or game management aspects
- A63F13/79—Game security or game management aspects involving player-related data, e.g. identities, accounts, preferences or play histories
- A63F13/798—Game security or game management aspects involving player-related data, e.g. identities, accounts, preferences or play histories for assessing skills or for ranking players, e.g. for generating a hall of fame
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- A63F13/00—Video games, i.e. games using an electronically generated display having two or more dimensions
- A63F13/70—Game security or game management aspects
- A63F13/79—Game security or game management aspects involving player-related data, e.g. identities, accounts, preferences or play histories
- A63F13/795—Game security or game management aspects involving player-related data, e.g. identities, accounts, preferences or play histories for finding other players; for building a team; for providing a buddy list
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- A63F2300/00—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game
- A63F2300/50—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game characterized by details of game servers
- A63F2300/55—Details of game data or player data management
- A63F2300/5546—Details of game data or player data management using player registration data, e.g. identification, account, preferences, game history
- A63F2300/5566—Details of game data or player data management using player registration data, e.g. identification, account, preferences, game history by matching opponents or finding partners to build a team, e.g. by skill level, geographical area, background, play style
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- A63F2300/00—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game
- A63F2300/50—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game characterized by details of game servers
- A63F2300/55—Details of game data or player data management
- A63F2300/5546—Details of game data or player data management using player registration data, e.g. identification, account, preferences, game history
- A63F2300/558—Details of game data or player data management using player registration data, e.g. identification, account, preferences, game history by assessing the players' skills or ranking
Abstract
The embodiment of the invention provides a data processing method, a data processing device, a server and a storage medium, wherein the method comprises the following steps: obtaining a first match result set, wherein the first match result set comprises a win-lose result of the M matches selected according to a first update granularity and a first user set participating in the M matches, and M is an integer greater than 1; acquiring a current capacity value of a target user, wherein the target user is any one user in the first user set; determining the updated ability value of the target user by using a target ability weight distribution method based on a webpage ranking algorithm, the first match result set and the current ability value of the target user; and determining a match matched with the target user according to the updated ability value of the target user, and calculating and updating the user ability accurately by the embodiment of the invention, thereby improving the flexibility of evaluating the user ability.
Description
Technical Field
The present invention relates to the field of computer technologies, and in particular, to a data processing method, an apparatus, a server, and a storage medium.
Background
With the development of sports competition and electronic game competition, the ability evaluation of the user in the competition is more and more important. Currently, the commonly used evaluation algorithms are the ranking algorithm (ELO), and the scoring algorithm developed by microsoft based on bayesian inference (TrueSkill). The methods are calculated and updated by taking a game as a unit according to the win-lose relationship of the game, so that the updating has problems, only the information of the current game is considered in each updating process, but the result information of other recent games of the user is not considered, the influence of the single game on the capability value of the user is overlarge, the fluctuation contingency of the capability value of the user is increased, and the method for evaluating the capability of the user is relatively fixed.
Disclosure of Invention
The embodiment of the invention provides a data processing method, a data processing device, a server and a storage medium, which can accurately calculate and update user capacity and improve the flexibility of user capacity evaluation.
A first aspect of an embodiment of the present invention provides a data processing method, including:
obtaining a first match result set, wherein the first match result set comprises a win-lose result of the M matches selected according to a first update granularity and a first user set participating in the M matches, and M is an integer greater than 1;
acquiring a current capacity value of a target user, wherein the target user is any one user in the first user set;
determining the updated ability value of the target user by using a target ability weight distribution method based on a webpage ranking algorithm, the first match result set and the current ability value of the target user;
and determining a match matched with the target user according to the updated ability value of the target user.
A second aspect of the embodiments of the present invention provides a data processing apparatus, including:
the device comprises an acquisition module, a comparison module and a comparison module, wherein the acquisition module is used for acquiring a first comparison result set, the first comparison result set comprises a win-lose result of M matches selected according to a first updating granularity and a first user set participating in the M matches, and M is an integer larger than 1;
the obtaining module is further configured to obtain a current ability value of a target user, where the target user is any one user in the first user set;
the determining module is used for determining the updated ability value of the target user by utilizing a target ability weight distribution method based on a webpage ranking algorithm, the first match result set and the current ability value of the target user;
the determining module is further configured to determine a match with the target user according to the updated ability value of the target user.
A third aspect of embodiments of the present invention provides a server, including a processor, a network interface, and a storage device, where the processor, the network interface, and the storage device are connected to each other, where the network interface is controlled by the processor to send and receive data, and the storage device is used to store a computer program, where the computer program includes program instructions, and the processor is configured to call the program instructions to execute the data processing method according to the first aspect.
A fourth aspect of the embodiments of the present invention provides a computer storage medium, in which program instructions are stored, and when the program instructions are executed, the computer storage medium is configured to implement the data processing method according to the first aspect.
In the embodiment of the invention, a server obtains a first match result set, wherein the first match result set comprises a win-lose result of M matches selected according to a first update granularity and a first user set participating in the M matches, M is an integer larger than 1, then obtains a current capacity value of a target user, the target user is any one user in the first user set, determines an updated capacity value of the target user by using a target capacity weight distribution method based on a webpage ranking algorithm, the first match result set and the current capacity value of the target user, and then determines a match matched with the target user according to the updated capacity value of the target user. The user ability can be accurately calculated and updated, and the flexibility of user ability evaluation is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1a is a schematic flow chart of a capability assessment algorithm provided by the prior art;
FIG. 1b is a schematic flow diagram of another capability assessment algorithm provided by the prior art;
FIG. 2 is a flow chart of a data processing method according to an embodiment of the present invention;
fig. 3a is a schematic diagram of a winning or losing result subgraph provided by the embodiment of the invention;
FIG. 3b is a diagram of another winning or losing result subgraph provided by the embodiment of the invention;
FIG. 4 is a diagram illustrating a method for assigning capacity weights based on exponential difference values according to an embodiment of the present invention;
FIG. 5 is a diagram illustrating a method for assigning capacity weights based on logistic regression differences according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a comparison of user ability values with actual rankings of games provided by embodiments of the invention;
FIG. 7 is a schematic diagram illustrating a comparison of chaos effects corresponding to different experimental algorithms according to an embodiment of the present invention;
FIG. 8 is a flow chart illustrating another data processing method according to an embodiment of the present invention;
FIG. 9 is a schematic structural diagram of a data processing apparatus according to an embodiment of the present invention;
fig. 10 is a schematic structural diagram of a server according to an embodiment of the present invention.
Detailed Description
With the research and progress of artificial intelligence technology, the artificial intelligence technology is developed and applied in a plurality of fields, such as common smart homes, smart wearable devices, virtual assistants, smart speakers, smart marketing, unmanned driving, automatic driving, unmanned aerial vehicles, robots, smart medical care, smart customer service, and the like.
Artificial Intelligence (AI) is a theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human Intelligence, perceive the environment, acquire knowledge and use the knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technique of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a manner similar to human intelligence. Artificial intelligence is the research of the design principle and the realization method of various intelligent machines, so that the machines have the functions of perception, reasoning and decision making.
The artificial intelligence technology is a comprehensive subject and relates to the field of extensive technology, namely the technology of a hardware level and the technology of a software level. The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.
Machine Learning (ML) is a multi-domain cross subject, and relates to multiple subjects such as probability theory, statistics, approximation theory, convex analysis and algorithm complexity theory. The special research on how a computer simulates or realizes the learning behavior of human beings so as to acquire new knowledge or skills and reorganize the existing knowledge structure to continuously improve the performance of the computer. Machine learning is the core of artificial intelligence, is the fundamental approach for computers to have intelligence, and is applied to all fields of artificial intelligence. Machine learning and deep learning generally include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and formal education learning. Among them, the data processing technology is an important technology for machine learning applications.
The data processing method provided by the embodiment of the invention relates to an artificial intelligence machine learning technology, and can obtain a first match result set, wherein the first match result set comprises a win-lose result of M matches selected according to a first update granularity and a first user set participating in the M matches, M is an integer larger than 1, then the current capacity value of a target user is obtained, the target user is any one user in the first user set, the updated capacity value of the target user is determined by using a target capacity weight distribution method based on a webpage ranking algorithm, the first match result set and the current capacity value of the target user, then the match matched with the target user is determined according to the updated capacity value of the target user, the user capacity can be accurately calculated and updated, and the flexibility of user capacity evaluation is improved. The following examples are intended to illustrate in particular:
the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Currently, Microsoft develops the TrueSkill ability evaluation algorithm, and assumes that the ability value S of a user follows Gaussian distribution N (S; mu, sigma), wherein mu represents the average of the ability of the user, sigma represents the fluctuation degree of the ability of the user, the actual expression p of the authority of the user follows the Gaussian distribution N (S; mu, beta), S represents the average of the ability expressed by the user, and beta represents the fluctuation degree of the ability expressed by the user. I (d ═ p) is expressed by an indicator function based on the difference between the actually exhibited capability levels of both parties to the battle1-p2) Predicting the future competition result based on the historical ability value, correcting the ability of the user according to the actual game result, adopting an updating method based on a factor graph (factor graph) in the whole process, and firstly predicting the competition result according to the historical ability value, wherein the ability values of the two parties are S respectively as shown in figures 1a and 1b1、S2Then according toIndication function representation I (d ═ p) of the difference between the actually exhibited capability levels of both parties in the battle1-p2) Then, the ability values of both the opponents are corrected by δ (r-sign (d)). By using the TrueSkill ability evaluation algorithm, the influence of the single game on the ability value of the user is too large, and the fluctuation contingency of the ability value of the user is increased.
Therefore, the embodiment of the invention adopts a webpage-based ranking algorithm (PageRank), sub-graphs are established according to the win-lose relationship between the users related to at least two games selected according to the specified updating granularity, weights are distributed to the users according to the sub-graphs by using different weight distribution algorithms, and then the ability value of the updated user is calculated according to the weights, so that the situation that the ability value of the user fluctuates too much due to single game is avoided, and the ability value of the updated user can be flexibly calculated.
The original PageRank algorithm is a method for calculating the importance of web pages according to the directions of hyperlinks between the web pages, and the basic idea of the algorithm is based on two assumptions:
(1) the quantity is assumed, the more the web pages pointing to one web page A, namely the more the in-degree is, the higher the web page importance is;
(2) the quality assumption is that the higher the web page quality pointing to the web page A is, the more important the A is, namely, the higher the weight is;
As can be seen from the above, the original PageRank algorithm averagely assigns the weight of each connection to other webpages in other directions, and in the embodiment of the present invention, the PageRank algorithm assigns the weight by calculating the ratio of the difference between the ability values of the winner and the winner on the direction side according to the current ability value of the user, rather than by average assignment, when evaluating the ability of the user.
Referring to fig. 2, fig. 2 is a flowchart illustrating a data processing method according to an embodiment of the present invention. The data processing method described in this embodiment includes the following steps:
201. and acquiring a first match result set, wherein the first match result set comprises the win and lose results of the M matches selected according to a first update granularity and a first user set participating in the M matches, and M is an integer larger than 1.
Specifically, the server obtains a first match result set including a win result and a lose result of the M matches selected according to a first update granularity and a first user set participating in the M matches, where M is an integer greater than 1. The first update granularity may be one day, half day, etc., and the M plays are the number of plays of all plays in the first update granularity. For example, if the first update granularity is half a day, which results in a win or a loss of 10 plays, then M is 10. Before the first competition result set is obtained, the competition results corresponding to the first update granularity need to be filtered out of AI users and on-hook users, so that the game results of at least two users are filtered, and the effectiveness of the obtained competition results is ensured.
202. And acquiring the current capability value of a target user, wherein the target user is any one user in the first user set.
203. And determining the updated ability value of the target user by using a target ability weight distribution method based on a webpage ranking algorithm, the first match result set and the current ability value of the target user.
Specifically, the server determines the updated ability value of the target user by using a target ability weight distribution method of a webpage ranking algorithm according to the first match result set and the current ability value of the target user.
In some feasible embodiments, the server determines the updated ability value of the target user by using the target ability weight assignment method based on the web page ranking algorithm, the first match result set and the current ability value of the target user, which may be determining an ability increment of the target user by using the target ability weight assignment method based on the web page ranking algorithm and the first match result set, and then calculating the updated ability value of the target user according to the ability increment of the target user and the current ability value.
204. And determining a match matched with the target user according to the updated ability value of the target user.
Specifically, the server matches users with the ability values similar to or different from those of the target users for competition according to the updated ability values of the target users, so that the experience and interestingness of the users are guaranteed.
In the embodiment of the invention, a server obtains a first match result set, wherein the first match result set comprises a win-lose result of M matches selected according to a first update granularity and a first user set participating in the M matches, M is an integer larger than 1, then obtains a current capacity value of a target user, the target user is any one user in the first user set, determines an updated capacity value of the target user by using a target capacity weight distribution method based on a webpage ranking algorithm, the first match result set and the current capacity value of the target user, and then determines a match matched with the target user according to the updated capacity value of the target user. The user ability can be accurately calculated and updated, and the flexibility of user ability evaluation is improved.
In some feasible embodiments, the updated ability value of the target user may be obtained in a unidirectional (i.e., forward) update manner, and specifically, the server may determine a forward ability increment of the target user by using a target ability weight assignment method based on a web page ranking algorithm and the first match result set, take the forward ability increment of the target user as the ability increment of the target user, and then calculate the updated ability value of the target user according to the ability increment of the target user and the current ability value.
In some possible embodiments, the obtaining the updated capability value of the target user may be performed by using a bidirectional (i.e. forward and reverse) updating manner, which specifically includes: the server determines a forward capacity increment and a reverse capacity increment of a target user by using a target capacity weight distribution method based on a webpage ranking algorithm and the first match result set; and determining the capacity increment of the target user according to the forward capacity increment and the reverse capacity increment, and then calculating to obtain the updated capacity value of the target user according to the capacity increment of the target user and the current capacity value.
In some feasible embodiments, the determining the capacity increment of the target user by using the target capacity weight distribution method based on the web page ranking algorithm and the first match result set may be that a win-lose result subgraph corresponding to the first match result set is constructed by using the target capacity weight distribution method based on the web page ranking algorithm, the win-lose result subgraph includes each user in the first user set having a win-lose relationship with the target user and a directing edge of the win-lose relationship, the current capacity value of each user is obtained, and the win-lose result subgraph is used for distributing the current capacity value of each user to obtain the total capacity value distributed by the target user; and taking the total capacity value allocated to the target user as the capacity increment of the target user.
The point side of the win-or-lose relationship refers to the point of the negative side to the win, and the construction of the win-or-lose result subgraph can be a one-way construction of the win-or-lose result subgraph among users or a two-way construction of the win-or-lose result subgraph. As shown in FIG. 3a, four users A, B, C, D previously constructed a one-way negative and positive result sub-graph, such as A, B, with only user B pointing to A; as shown in FIG. 3b, four users A, B, C, D previously constructed bi-directional win-loss result subgraphs, such as the bi-directional connection between A, B.
For example, based on fig. 3a, a unidirectional win-or-lose result sub-graph is constructed among the four users A, B, C, D, and the sum Δ of the capacity values scored by the target user a is calculated according to the constructed unidirectional win-or-lose result sub-graph, taking a as an exampleprI.e. target user capacity deltaprThen obtain the current ability value of A as PRoldThen the updated capability value of the target user A is
PRnew＝PRold+Δpr
For another example, based on fig. 3b, taking a as an example, the users A, B, C, D construct a bidirectional win-or-lose result sub-graph before, and calculate the sum Δ of forward-divided capacity values of the target user a according to the constructed bidirectional win-or-lose result sub-graphpr positiveSum of capability values Δ divided in the forward directionpr negativeObtaining the target user A scoreSum of capability values Δpr＝Δpr positive-Δpr negativeI.e. target user capacity deltaprThen obtain the current ability value of A as PRoldIf the updated target user A has the capability value PRnew＝PRold+Δpr。
In some feasible embodiments, the winning or negating result subgraph is used for distributing the current ability values of the users, and the total ability value distributed by the target user can be obtained by obtaining the ability difference value of the current ability values between the target user and the users by using the directional edge of the winning or negating relation in the winning or negating result subgraph, and the ability distribution proportion of the users to the target user is calculated according to the ability difference value; then calculating the ability value of each user distributed to the target user by using the current ability value and the ability distribution proportion of each user; and determining the total capacity value allocated to the target user according to the capacity value allocated to the target user by the current capacity value of each user. Wherein, the directional edge of the victory-negative relationship refers to that the negative side points to the victory, and the total capacity value allocated to the target user is the increment of the capacity value of the target user.
In some possible embodiments, the method for assigning a target ability weight with the best effect may be found from at least two methods for assigning an ability weight based on a web page ranking algorithm, and specifically may include: the server obtains a second competition result set, wherein the second competition result set comprises a win-lose result of N competitions selected according to a second updating granularity and a second user set participating in the N competitions, and N is an integer larger than 1; determining at least two updated capacity values of each user included in the second user set by utilizing at least two capacity weight distribution methods based on a webpage ranking algorithm and the second competition result set; matching the users participating in the game for each user by using at least two updated capability values of each user in the S game determined according to the third update granularity, wherein S is an integer larger than 1; and determining a target ability weight distribution method from at least two ability weight distribution methods according to the competition result of the S game.
Before obtaining the second match result set, the server needs to filter the AI users and the on-hook users from the match results in the second update granularity, where the second update granularity may be one day, half day, and so on, and the N matches are the number of all matches in the second update granularity. For example, if the second update granularity is half a day, which yields a win or a lose result of 10 plays, then N is 10.
The at least two capacity allocation methods may include a capacity weight allocation method based on an exponential difference value and a capacity weight allocation method based on a logistic regression difference value, and the capacity weight allocation principle satisfies: difference between the winner and negation ability values Δpr_win-pr_lostThe larger the E (-infinity, + ∞) the less weight value, and conversely Δpr_win-pr_lostThe smaller the weight value, the more the weight value.
For example, based on fig. 3b, the ability weight assignment method using exponential difference can obtain the user ability increment:
using the difference of the user's competence Δpr_win-pr_lostTo assign weights in the form of exponential function values.
Wherein PR (A) represents the increment of the capacity (corresponding to the above-mentioned. DELTA.)pr) And l (B) represents the ability allocation ratio of the above-described ability value of the user B to the user a.
Further, the capacity increment of a certain user is:
PRin_jrepresenting the value of the winner's incremental ability, PRin_iRepresenting the current ability value of the winner, PRoutRepresenting the current capability value of a negative party, n representing the output number of the current node, and j representing the jth output.
As shown in fig. 4, a fingerIn the diagram of the number difference capacity weight assignment PageRank method, the horizontal axis x represents the difference between the positive and negative capacities, i.e. x equals to PRin_j-PRout。
For another example, based on fig. 3b, the ability weight distribution method using the logistic regression difference can obtain the user ability increment:
using the difference of the user's competence Δpr_win-pr_lostTo assign weights in the form of logical distribution values.
Wherein PR (A) represents the increment of the capacity (corresponding to the above-mentioned. DELTA.)pr) And l (B) represents the ability allocation ratio of the above-described ability value of the user B to the user a.
Further, the capacity increment of a certain user is:
as shown in fig. 5, in the schematic diagram of the exponential difference capability weight assignment PageRank method, the horizontal axis x represents the difference between the positive and negative capabilities, i.e., x is PRin_j-PRout。
And determining at least two updated ability values of each user included in the second user set by using at least two ability weight distribution methods based on the webpage ranking algorithm and the second competition result set, wherein the at least two updated ability values are the ability values obtained by updating after calculating the corresponding increment after the weights are distributed by the two algorithms, and the specific updating mode refers to the embodiment.
In some possible embodiments, the server may determine the target ability weight distribution method from the at least two ability weight distribution methods according to the match result of the S-round match by obtaining an evaluation value of each of the at least two ability weight distribution methods according to the match result of the S-round match, where the evaluation value is a chaos degree evaluation index, and determining the target ability weight distribution method from the at least two ability weight distribution methods according to the evaluation value.
The chaos degree evaluation index represents the proportion of the chaos condition of the actual ability ranking and the actual competition result ranking of the user in a competition, and the specific calculation formula of the chaos degree evaluation index is as follows:
where n represents the number of users in the match result, ranki、rankjRepresenting the user's actual result ranking, skilli、skilljRepresenting the user's actual ability ranking.
In order to verify the effect situation of the embodiment of the invention, the latest capacity value of the user is obtained by calculating the commonly matched game data of a certain game in the time interval of 20181201 and 20181223, and then the field average chaos is calculated by adopting the actual game result ranking and the user capacity value ranking of 20181224 to verify the accuracy of various capacity calculation methods, wherein the total number of games in the verification set is 2587335, and the total number of users is 8723067. The chaos degree represents the proportion of chaos in a game where the user's actual ability is ranked and the actual game results are ranked. As shown in FIG. 6, the comparison between the ability value of the user and the actual ranking of the game reveals that the ability value is confused with the actual ranking of the game. And calculating the chaos of each game result according to each game result of the verification set, so that the average value of the field average chaos can be calculated, and the smaller the field average chaos is, the more accurate the user capacity value is represented.
The result of calculating the field average chaos mean values of various ability evaluation algorithms according to the chaos formula is shown in fig. 7, and it can be seen that the chaos of the ability evaluation algorithm of the index difference value updated bidirectionally is the lowest among the various algorithms of PageRank, but differs from the chaos of TrueSkill by 6%, but PageRank can be dynamically adjusted according to the update granularity, and the ability value calculation is more flexible.
In some possible embodiments, the server adjusts the user's capability value over a period of time. Since the user's ability is not always constant, but is directly related to the user's proficiency in playing the game, the more proficient the user is in the game in the near future, the higher the ability value that can be expressed. Therefore, we are updating the current ability value of the user (e.g. the current ability value is PR (t)0) Time of the last game play (e.g., t) of the user is calculated0) From the present time (t), the interval of two times (Δ ═ t-t)0) And then adjusts the ability value of the user. That is, the adjusted ability value of the user is PR (t) ═ PR (t)0)×e-kΔWherein PR (t)0) Representing the capability value before adjustment, k being the adjustable attenuation coefficient.
In some feasible embodiments, as shown in fig. 8, AI on-hook users and abnormal result bureaus are filtered according to historical competition results, winning and losing result subgraphs are constructed according to day granularity accumulation, and then the ability values are updated according to different PageRank methods to obtain the ability values of the current users; and filtering an AI on-hook user and an abnormal result by combining the competition result of the current day, calculating the chaos according to the actual result ranking and the current capability value ranking, and matching the users according to the latest capability value and the matching strategy on line.
Referring to fig. 9, fig. 9 is a schematic structural diagram of a data processing apparatus according to an embodiment of the present invention. The data processing apparatus described in this embodiment includes:
an obtaining module 901, configured to obtain a first match result set, where the first match result set includes a win/lose result of an M-game match selected according to a first update granularity and a first user set participating in the M-game match, and M is an integer greater than 1;
the obtaining module 901 is further configured to obtain a current ability value of a target user, where the target user is any user in the first user set;
a determining module 902, configured to determine an updated ability value of the target user by using a target ability weight assignment method based on a web page ranking algorithm, the first match result set, and the current ability value of the target user;
the determining module 902 is further configured to determine a match matching the target user according to the updated ability value of the target user.
Optionally, the determining module 902 is specifically configured to:
determining the capacity increment of the target user by utilizing a target capacity weight distribution method based on a webpage ranking algorithm and the first match result set;
and calculating to obtain the updated ability value of the target user according to the ability increment of the target user and the current ability value.
Optionally, the apparatus comprises: a build module 903, wherein:
the constructing module 903 is configured to construct a winning or losing result subgraph corresponding to the first match result set by using a target ability weight distribution method based on a web page ranking algorithm, where the winning or losing result subgraph includes users in the first user set who have a winning or losing relationship with the target user and a directing edge of the winning or losing relationship;
the obtaining module 901 is further configured to obtain the current ability value of each user, and allocate the current ability value of each user by using the win-lose result subgraph to obtain the total amount of ability values allocated to the target user;
the determining module 902 is further configured to use a total amount of the capacity values allocated to the target user as the capacity increment of the target user.
Optionally, the determining module 902 is specifically configured to:
determining a forward capacity increment and a reverse capacity increment of the target user by utilizing a target capacity weight distribution method based on a webpage ranking algorithm and the first match result set;
and determining the capacity increment of the target user according to the forward capacity increment and the reverse capacity increment.
Optionally, the determining module 902 is specifically configured to:
acquiring the capability difference value of the current capability value between the target user and each user by using the directional edge of the winning or losing relationship in the winning or losing result subgraph;
calculating the capacity distribution proportion of each user to the target user according to the capacity difference;
calculating the ability value of each user distributed to the target user according to the current ability value of each user and the ability distribution proportion;
and determining the total capacity value allocated to the target user according to the capacity value allocated to the target user by the current capacity value of each user.
Optionally, the apparatus further comprises: a matching module 904, wherein:
the obtaining module 901 is further configured to obtain a second match result set, where the second match result set includes a win/lose result of N matches selected according to a second update granularity and a second user set participating in the N matches, where N is an integer greater than 1;
the determining module 902 is further configured to determine at least two updated ability values of each user included in the second set of users by using at least two ability weight assignment methods based on the web page ranking algorithm and the second set of game results;
the matching module 904 is configured to match a participating user for each user by using the at least two updated ability values of each user in an S round match determined according to a third update granularity, where S is an integer greater than 1;
the determining module 902 is further configured to determine a target ability weight distribution method from the at least two ability weight distribution methods according to a match result of the S game.
Optionally, the determining module 902 is specifically configured to:
obtaining an evaluation value of each ability weight distribution method in the at least two ability weight distribution methods according to a competition result of the S game, wherein the evaluation value is a chaos degree evaluation index;
and determining a target ability weight distribution method from the at least two ability weight distribution methods according to the evaluation value.
It can be understood that the functions of the functional modules of the data processing apparatus in this embodiment may be specifically implemented according to the method in the foregoing method embodiment, and the specific implementation process may refer to the relevant description of the foregoing method embodiment, which is not described herein again.
In this embodiment, the obtaining module 901 obtains a first match result set, where the first match result set includes a win/lose result of M matches selected according to a first update granularity and a first user set participating in the M matches, where M is an integer greater than 1, the obtaining module 901 obtains a current capability value of a target user, the target user is any one of the first user set, then the determining module 902 determines an updated capability value of the target user by using a target capability weight distribution method based on a web page ranking algorithm, the first match result set, and the current capability value of the target user, and the determining module 902 determines a match matched with the target user according to the updated capability value of the target user, so that the user capability can be accurately calculated and updated, and the flexibility of evaluating the user capability is improved.
Referring to fig. 10, fig. 10 is a schematic structural diagram of a server according to an embodiment of the present invention. The server described in this embodiment includes: the method comprises the following steps: a processor 1001, a network interface 1002, and a memory 1003. The processor 1001, the network interface 1002, and the memory 1003 may be connected by a bus or other methods, and the embodiment of the present invention is exemplified by connection via a bus.
The processor 1001 (or Central Processing Unit (CPU)) is a computing core and a control core of the server. The network interface 1002 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI, mobile communication interface, etc.), controlled by the processor 1001 for transceiving data. The Memory 1003(Memory) is a Memory device of the server, and stores programs and data. It is understood that the memory 1003 may be a high-speed RAM memory, or may be a non-volatile memory (non-volatile memory), such as at least one disk memory; optionally, at least one memory device may be located remotely from the processor 1001. Memory 1003 provides storage space that stores the operating system and executable program code for the server, which may include, but is not limited to: windows system (an operating system), Linux system (an operating system), etc., which are not limited in this regard.
In the embodiment of the present invention, the processor 1001 executes the executable program code in the memory 1003 to perform the following operations:
obtaining a first match result set, wherein the first match result set comprises a win-lose result of the M matches selected according to a first update granularity and a first user set participating in the M matches, and M is an integer greater than 1;
acquiring a current capacity value of a target user, wherein the target user is any one user in the first user set;
determining the updated ability value of the target user by using a target ability weight distribution method based on a webpage ranking algorithm, the first match result set and the current ability value of the target user;
and determining a match matched with the target user according to the updated ability value of the target user.
Optionally, the processor 1001 is configured to:
determining the capacity increment of the target user by utilizing a target capacity weight distribution method based on a webpage ranking algorithm and the first match result set;
and calculating to obtain the updated ability value of the target user according to the ability increment of the target user and the current ability value.
Optionally, the processor 1001 is configured to:
constructing a winning and losing result subgraph corresponding to the first competition result set by using a target ability weight distribution method based on a webpage ranking algorithm, wherein the winning and losing result subgraph comprises users in the first user set, which have winning and losing relations with the target users, and directing edges of the winning and losing relations;
acquiring the current capacity value of each user, and distributing the current capacity value of each user by using the victory or defeat result subgraph to obtain the total capacity value distributed by the target user;
and taking the total capacity value allocated to the target user as the capacity increment of the target user.
Optionally, the processor 1001 is configured to:
determining a forward capacity increment and a reverse capacity increment of the target user by utilizing a target capacity weight distribution method based on a webpage ranking algorithm and the first match result set;
and determining the capacity increment of the target user according to the forward capacity increment and the reverse capacity increment.
Optionally, the processor 1001 is configured to:
acquiring the capability difference value of the current capability value between the target user and each user by using the directional edge of the winning or losing relationship in the winning or losing result subgraph;
calculating the capacity distribution proportion of each user to the target user according to the capacity difference;
calculating the ability value of each user distributed to the target user according to the current ability value of each user and the ability distribution proportion;
and determining the total capacity value allocated to the target user according to the capacity value allocated to the target user by the current capacity value of each user.
Optionally, the processor 1001 is further configured to:
acquiring a second competition result set, wherein the second competition result set comprises a win-lose result of N competitions selected according to a second updating granularity and a second user set participating in the N competitions, and N is an integer larger than 1;
determining at least two updated ability values of each user included in the second set of users by utilizing at least two ability weight distribution methods based on the webpage ranking algorithm and the second competition result set;
matching each user with a participating user in an S game determined according to a third update granularity by using at least two updated capability values of each user, wherein S is an integer greater than 1;
and determining a target ability weight distribution method from the at least two ability weight distribution methods according to the competition result of the S game.
Optionally, the processor 1001 is further configured to:
obtaining an evaluation value of each ability weight distribution method in the at least two ability weight distribution methods according to a competition result of the S game, wherein the evaluation value is a chaos degree evaluation index;
and determining a target ability weight distribution method from the at least two ability weight distribution methods according to the evaluation value.
In a specific implementation, the processor 1001, the network interface 1002, and the memory 1003 described in the embodiment of the present invention may execute an implementation manner described in a flow of a data processing method provided in the embodiment of the present invention, and may also execute an implementation manner described in a data processing apparatus provided in the embodiment of the present invention, which is not described herein again.
The embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored, where the computer program includes program instructions, and when the program instructions are executed by a processor, the steps performed in the above data processing embodiments may be performed.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (8)
1. A method of data processing, comprising:
acquiring a second competition result set, wherein the second competition result set comprises a win-lose result of N competitions selected according to a second updating granularity and a second user set participating in the N competitions, and N is an integer larger than 1;
determining at least two updated ability values of each user included in the second user set by utilizing at least two ability weight distribution methods based on a webpage ranking algorithm and the second competition result set;
matching each user with a participating user in an S game determined according to a third update granularity by using at least two updated capability values of each user, wherein S is an integer greater than 1;
determining a target ability weight distribution method from the at least two ability weight distribution methods according to a competition result of the S game;
obtaining a first match result set, wherein the first match result set comprises a win-lose result of the M matches selected according to a first update granularity and a first user set participating in the M matches, and M is an integer greater than 1;
acquiring a current capacity value of a target user, wherein the target user is any one user in the first user set;
determining the capacity increment of the target user by using the target capacity weight distribution method and the first match result set;
calculating to obtain the updated ability value of the target user according to the ability increment of the target user and the current ability value;
and determining a match matched with the target user according to the updated ability value of the target user.
2. The method of claim 1, wherein determining the capacity increment of the target user by using a target capacity weight assignment method based on a web page ranking algorithm and the first match result set comprises:
constructing a winning and losing result subgraph corresponding to the first competition result set by using a target ability weight distribution method based on a webpage ranking algorithm, wherein the winning and losing result subgraph comprises users in the first user set, which have winning and losing relations with the target users, and directing edges of the winning and losing relations;
acquiring the current capacity value of each user, and distributing the current capacity value of each user by using the victory or defeat result subgraph to obtain the total capacity value distributed by the target user;
and taking the total capacity value allocated to the target user as the capacity increment of the target user.
3. The method of claim 1, wherein determining the capacity increment of the target user by using a target capacity weight assignment method based on a web page ranking algorithm and the first match result set comprises:
determining a forward capacity increment and a reverse capacity increment of the target user by utilizing a target capacity weight distribution method based on a webpage ranking algorithm and the first match result set;
and determining the capacity increment of the target user according to the forward capacity increment and the reverse capacity increment.
4. The method of claim 2, wherein the assigning the current ability values of the users by using the win-lose result subgraph to obtain the total amount of ability values assigned by the target user comprises:
acquiring the capability difference value of the current capability value between the target user and each user by using the directional edge of the winning or losing relationship in the winning or losing result subgraph;
calculating the capacity distribution proportion of each user to the target user according to the capacity difference;
calculating the ability value of each user distributed to the target user according to the current ability value of each user and the ability distribution proportion;
and determining the total capacity value allocated to the target user according to the capacity value allocated to the target user by the current capacity value of each user.
5. The method according to claim 1, wherein the determining a target ability weight assignment method from the at least two ability weight assignment methods according to the game result of the S round game comprises:
obtaining an evaluation value of each ability weight distribution method in the at least two ability weight distribution methods according to a competition result of the S game, wherein the evaluation value is a chaos degree evaluation index;
and determining a target ability weight distribution method from the at least two ability weight distribution methods according to the evaluation value.
6. An apparatus for data processing, comprising:
an obtaining module, configured to obtain a second match result set, where the second match result set includes a win/lose result of N matches selected according to a second update granularity and a second user set participating in the N matches, and N is an integer greater than 1;
the determining module is used for determining at least two updated capacity values of each user included in the second user set by utilizing at least two capacity weight distribution methods based on a webpage ranking algorithm and the second competition result set;
the determining module is further configured to match a participating user for each user by using the at least two updated ability values of each user in an S round match determined according to a third update granularity, where S is an integer greater than 1;
the determining module is further configured to determine a target ability weight distribution method from the at least two ability weight distribution methods according to a match result of the S game;
the obtaining module is further configured to obtain a first match result set, where the first match result set includes a win/lose result of the M matches selected according to a first update granularity and a first user set participating in the M matches, and M is an integer greater than 1;
the obtaining module is further configured to obtain a current ability value of a target user, where the target user is any one user in the first user set;
the determining module is further configured to determine a capacity increment of the target user by using the target capacity weight assignment method and the first match result set; calculating to obtain the updated ability value of the target user according to the ability increment of the target user and the current ability value;
the determining module is further configured to determine a match with the target user according to the updated ability value of the target user.
7. A server, comprising a processor, a network interface and a storage device, wherein the processor, the network interface and the storage device are connected with each other, wherein the network interface is controlled by the processor for transceiving data, the storage device is used for storing a computer program, the computer program comprises program instructions, and the processor is configured to call the program instructions to execute the data processing method according to any one of claims 1 to 5.
8. A computer storage medium having stored thereon program instructions for performing, when executed, a method of data processing according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911037287.4A CN110772797B (en) | 2019-10-29 | 2019-10-29 | Data processing method, device, server and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911037287.4A CN110772797B (en) | 2019-10-29 | 2019-10-29 | Data processing method, device, server and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110772797A CN110772797A (en) | 2020-02-11 |
| CN110772797B true CN110772797B (en) | 2021-09-28 |
Family
ID=69387266
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911037287.4A Active CN110772797B (en) | 2019-10-29 | 2019-10-29 | Data processing method, device, server and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110772797B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112402982B (en) * | 2020-02-13 | 2022-12-23 | 上海哔哩哔哩科技有限公司 | User cheating behavior detection method and system based on machine learning |
| CN111659126B (en) * | 2020-07-08 | 2023-03-03 | 腾讯科技（深圳）有限公司 | Distribution method, device, server, terminal and storage medium of matching process |
| CN111803959B (en) * | 2020-07-21 | 2022-05-31 | 腾讯科技（深圳）有限公司 | Object control method, device, game server and medium |
| CN113827985B (en) * | 2021-08-12 | 2022-06-28 | 中投国信(北京)科技发展有限公司 | Big data service information evaluation method and system |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101313322A (en) * | 2005-11-21 | 2008-11-26 | 微软公司 | Team matching |
| CN101379523A (en) * | 2006-02-10 | 2009-03-04 | 微软公司 | Determining relative skills of players |
| JP2009172414A (en) * | 2009-05-11 | 2009-08-06 | Ace Denken Co Ltd | Game machine |
| CN102682196A (en) * | 2012-04-13 | 2012-09-19 | 青岛美天网络科技有限公司 | Method and system for graphical display |
| CN104436656A (en) * | 2014-12-24 | 2015-03-25 | 深圳市腾讯计算机系统有限公司 | Method and device for matching users in game application |
| JP2015109901A (en) * | 2013-12-06 | 2015-06-18 | 株式会社カプコン | Game program and game device |
| JP2016087015A (en) * | 2014-10-31 | 2016-05-23 | 株式会社バンダイナムコエンターテインメント | Server system and program |
| US9440152B2 (en) * | 2013-05-22 | 2016-09-13 | Clip Engine LLC | Fantasy sports integration with video content |
| CN106844778A (en) * | 2017-03-07 | 2017-06-13 | 北京齐尔布莱特科技有限公司 | A kind of data processing method, device and the webserver |
| CN107077708A (en) * | 2014-09-15 | 2017-08-18 | 洪国际有限公司 | Computer program for the method and apparatus of online competition schedule and for implementing online competition schedule |
| CN108379846A (en) * | 2018-03-16 | 2018-08-10 | 深圳冰川网络股份有限公司 | A kind of player's On-line matching method and system for existence class game |
| CN108392828A (en) * | 2018-03-16 | 2018-08-14 | 深圳冰川网络股份有限公司 | A kind of player's On-line matching method and system for the game of MOBA classes |
| CN108786116A (en) * | 2018-04-24 | 2018-11-13 | 网易（杭州）网络有限公司 | Team matching method, electronic equipment and storage medium in battle game |
| CN109032671A (en) * | 2018-06-25 | 2018-12-18 | 电子科技大学 | A kind of distributed deep learning method and system based on data parallel strategy |
| CN109453524A (en) * | 2018-11-14 | 2019-03-12 | 腾讯科技（深圳）有限公司 | A kind of method of object matching, the method for model training and server |
| CN109513215A (en) * | 2018-11-23 | 2019-03-26 | 腾讯科技（深圳）有限公司 | A kind of object matching method, model training method and server |
| CN109603159A (en) * | 2018-12-18 | 2019-04-12 | 网易（杭州）网络有限公司 | Match the method, apparatus and system of game player |
| CN110152304A (en) * | 2018-05-07 | 2019-08-23 | 腾讯科技（深圳）有限公司 | Determination method and device, storage medium and the electronic device of triumph value |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070112706A1 (en) * | 2005-01-24 | 2007-05-17 | Microsoft Corporation | Handicapping in a Bayesian skill scoring framework |
| JP2019050938A (en) * | 2017-09-13 | 2019-04-04 | 任天堂株式会社 | Information processing system, information processing device, information processing server, information processing program, and information processing method |
-
2019
- 2019-10-29 CN CN201911037287.4A patent/CN110772797B/en active Active
Patent Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101313322A (en) * | 2005-11-21 | 2008-11-26 | 微软公司 | Team matching |
| CN101379523A (en) * | 2006-02-10 | 2009-03-04 | 微软公司 | Determining relative skills of players |
| JP2009172414A (en) * | 2009-05-11 | 2009-08-06 | Ace Denken Co Ltd | Game machine |
| CN102682196A (en) * | 2012-04-13 | 2012-09-19 | 青岛美天网络科技有限公司 | Method and system for graphical display |
| US9440152B2 (en) * | 2013-05-22 | 2016-09-13 | Clip Engine LLC | Fantasy sports integration with video content |
| JP2015109901A (en) * | 2013-12-06 | 2015-06-18 | 株式会社カプコン | Game program and game device |
| CN107077708A (en) * | 2014-09-15 | 2017-08-18 | 洪国际有限公司 | Computer program for the method and apparatus of online competition schedule and for implementing online competition schedule |
| JP2016087015A (en) * | 2014-10-31 | 2016-05-23 | 株式会社バンダイナムコエンターテインメント | Server system and program |
| CN104436656A (en) * | 2014-12-24 | 2015-03-25 | 深圳市腾讯计算机系统有限公司 | Method and device for matching users in game application |
| CN106844778A (en) * | 2017-03-07 | 2017-06-13 | 北京齐尔布莱特科技有限公司 | A kind of data processing method, device and the webserver |
| CN108379846A (en) * | 2018-03-16 | 2018-08-10 | 深圳冰川网络股份有限公司 | A kind of player's On-line matching method and system for existence class game |
| CN108392828A (en) * | 2018-03-16 | 2018-08-14 | 深圳冰川网络股份有限公司 | A kind of player's On-line matching method and system for the game of MOBA classes |
| CN108786116A (en) * | 2018-04-24 | 2018-11-13 | 网易（杭州）网络有限公司 | Team matching method, electronic equipment and storage medium in battle game |
| CN110152304A (en) * | 2018-05-07 | 2019-08-23 | 腾讯科技（深圳）有限公司 | Determination method and device, storage medium and the electronic device of triumph value |
| CN109032671A (en) * | 2018-06-25 | 2018-12-18 | 电子科技大学 | A kind of distributed deep learning method and system based on data parallel strategy |
| CN109453524A (en) * | 2018-11-14 | 2019-03-12 | 腾讯科技（深圳）有限公司 | A kind of method of object matching, the method for model training and server |
| CN109513215A (en) * | 2018-11-23 | 2019-03-26 | 腾讯科技（深圳）有限公司 | A kind of object matching method, model training method and server |
| CN109603159A (en) * | 2018-12-18 | 2019-04-12 | 网易（杭州）网络有限公司 | Match the method, apparatus and system of game player |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110772797A (en) | 2020-02-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110772797B (en) | Data processing method, device, server and storage medium | |
| CN108873936B (en) | Autonomous aircraft formation method based on potential game | |
| CN111401556A (en) | Selection method of opponent type imitation learning winning incentive function | |
| CN105637540A (en) | Methods and apparatus for reinforcement learning | |
| CN108416615B (en) | Resource allocation method based on tree | |
| WO2016128960A1 (en) | System and method for retaining a strategy video game player by predicting the player game satisfaction using player game behavior data | |
| CN111176820A (en) | Deep neural network-based edge computing task allocation method and device | |
| Sekhavat | MPRL: Multiple-Periodic Reinforcement Learning for difficulty adjustment in rehabilitation games | |
| Cui et al. | Using social emotional optimization algorithm to direct orbits of chaotic systems | |
| CN115018017B (en) | Multi-agent credit allocation method, system and equipment based on ensemble learning | |
| CN113962390A (en) | Method for constructing diversified search strategy model based on deep reinforcement learning network | |
| CN108876123A (en) | A kind of teaching interference method and device | |
| Smyrnakis et al. | Dynamic opponent modelling in fictitious play | |
| CN110472792A (en) | A kind of route optimizing method for logistic distribution vehicle based on discrete bat algorithm | |
| CN110188123A (en) | User matching method and equipment | |
| CN116302569B (en) | Resource partition intelligent scheduling method based on user request information | |
| CN110827075A (en) | Data processing method and device, server and computer readable storage medium | |
| CN113162861A (en) | Communication method, device and system based on model training | |
| CN106294395B (en) | A kind of method and device of task processing | |
| CN109543725A (en) | A kind of method and device obtaining model parameter | |
| EP3561742A1 (en) | Test and training data | |
| US11478716B1 (en) | Deep learning for data-driven skill estimation | |
| CN114816755A (en) | Scheduling method, scheduling device, processing core, electronic device and readable medium | |
| CN103955802A (en) | Participatory data perception system and method for promoting data contribution by users | |
| Kretchmar | Reinforcement learning algorithms for homogenous multi-agent systems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 40021122 Country of ref document: HK |
|
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |