CN110197191A

CN110197191A - Electronic game recommended method

Info

Publication number: CN110197191A
Application number: CN201810929412.1A
Authority: CN
Inventors: 王星雅
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2018-08-15
Filing date: 2018-08-15
Publication date: 2019-09-03
Anticipated expiration: 2038-08-15
Also published as: CN110197191B

Abstract

The present invention relates to a kind of electronic game recommended methods, belong to Internet technical field.The electronic game recommended method includes: to constitute sample set for the associated electronic game group of each sample objects as sample；Wherein, the quantity of electronic game is greater than 1 in the electronic game group；The feature vector of electronic game in each electronic game group is obtained according to sample set training；It determines the feature vector of target object, and calculates the similarity of the feature vector of the target object and the feature vector of an electronic game；The target object is determined for the preference of electronic game described in this, if target object is greater than preset threshold for the preference of electronic game described in this, it is determined that the electronic game is target electronic game according to the similarity；And give the determining target electronic game recommdation to the target object.The accuracy of electronic game recommendation can be improved in the present invention.

Description

Electronic game recommendation method

Technical Field

The invention belongs to the technical field of internet, and particularly relates to an electronic game recommendation method.

Background

Along with the continuous improvement of the living standard of people, the entertainment requirements of people are higher and higher, and more people like playing electronic games in spare time. With the rapid development of internet technology, various electronic games are distributed in the network. To enhance the user experience, it is often necessary to recommend to the user an electronic game that is of interest to the user.

However, in the prior art, it often happens that the recommended electronic game is not the electronic game in which the user is interested, and thus a good electronic game recommendation effect is not achieved, so that the recommendation resource is wasted, and meanwhile, a poor user experience is brought to the user.

Therefore, how to recommend the electronic game to the user more accurately becomes a problem to be solved.

It is to be noted that the information disclosed in the above background section is only for enhancement of understanding of the background of the present invention and therefore may include information that does not constitute prior art known to a person of ordinary skill in the art.

Disclosure of Invention

The present invention is directed to a method, an apparatus, an electronic device and a computer-readable storage medium for recommending an electronic game, which overcome, at least to some extent, the problems of poor accuracy in recommending an electronic game due to the limitations and disadvantages of the related art.

According to a first aspect of the present invention, there is provided an electronic game recommendation method comprising:

respectively taking the electronic game groups associated with the sampling objects as samples to form a sample set; wherein the number of electronic games in the set of electronic games is greater than 1;

training according to the sample set to obtain the feature vectors of the electronic games in each electronic game group;

determining a feature vector of a target object, and calculating the similarity between the feature vector of the target object and the feature vector of the electronic game;

determining the preference degree of the target object to the electronic game according to the similarity, and if the preference degree of the target object to the electronic game is greater than a preset threshold value, determining that the electronic game is the target electronic game;

recommending the determined target electronic game to the target object.

In an exemplary embodiment of the invention, the sample object associated video game set includes one or more of: the electronic game which is contacted by the sampling object in a preset time, the historical registration electronic game of the sampling object, the electronic game which has the active time of the sampling object meeting a preset time length, and the electronic game which has the consumption record of the sampling object.

In an exemplary embodiment of the present invention, the training to obtain the feature vector of the electronic game in each electronic game group according to the sample set includes:

determining an initial vector for each of the electronic games in the sample set, respectively;

and training the initial vector of each electronic game to obtain the feature vector of each electronic game.

In an exemplary embodiment of the present invention, the training of the initial vector of each of the electronic games separately comprises:

training the initial vector of each electronic game based on a probability regression function.

In an exemplary embodiment of the present invention, training the initial vector of each of the electronic games separately comprises:

respectively selecting each sample as a current sample, and respectively selecting each electronic game in the current sample as a current electronic game;

combining the current electronic game and a positive sample element into a first sample;

for the first example, with a label as 1 as a target, performing first training on the current electronic game and the initial vector of the positive sample element through the probability regression function;

wherein the positive sample element is an electronic game in the current sample other than the current electronic game.

In an exemplary embodiment of the present invention, training the initial vector of each of the electronic games separately further comprises:

combining the current electronic game and a negative sample element into a second sample;

for the second example, taking a label as 0 as a target, and performing second training on the current electronic game and the initial vector of the negative sample element through the probability regression function;

wherein the negative sample element is an electronic game in a sample other than the current sample in the set of samples.

In an exemplary embodiment of the invention, the method further comprises:

clustering all the electronic games according to the feature vectors of all the electronic games;

judging whether the characteristic vector of each electronic game reaches a target or not according to the clustering result;

and when the feature vector of each electronic game is judged not to reach the target, training according to the sample set again to obtain the feature vector of the electronic game in each electronic game group.

In an exemplary embodiment of the invention, the method further comprises:

acquiring the operation behavior of the target object on the target electronic game;

judging whether the characteristic vector of each electronic game reaches a target or not according to the operation behavior of the target object;

In an exemplary embodiment of the present invention, determining a degree of preference of a target object for the electronic game according to each of the similarities includes:

representing the preference degree of the target object for the electronic game through the maximum value in the similarity; or

Representing the preference degree of the target object for the electronic game through the average value of the similarity degrees; or

And expressing the preference degree of the target object to the electronic game through the weighted average of the similarity degrees.

In an exemplary embodiment of the present invention, determining a feature vector of a target object includes:

respectively taking the feature vectors of the electronic games related to the target object as the feature vectors of the target object; or,

calculating all the feature vectors of the electronic game related to the target object to obtain the feature vector of the target object; or,

and determining related objects having a preset relation with the target object, and determining the feature vector of the target object according to the feature vectors of all the electronic games related to the related objects.

According to a second aspect of the present invention, there is provided an electronic game recommendation apparatus comprising:

the sample set constructing module is used for respectively taking the electronic game groups associated with the sampling objects as samples to form a sample set; wherein the number of electronic games in the set of electronic games is greater than 1;

the characteristic determining module is used for obtaining the characteristic vector of the electronic game in each electronic game group according to the sample set training;

the recommendation judging module is used for determining a feature vector of a target object, calculating the similarity between the feature vector of the target object and the feature vector of the electronic game, determining the preference degree of the target object for the electronic game according to the similarity, and determining the electronic game as the target electronic game if the preference degree of the target object for the electronic game is greater than a preset threshold value; and

and the game recommending module is used for recommending the determined target electronic game to the target object.

In an exemplary embodiment of the invention, the feature determination module includes:

a vector initialization unit for determining an initial vector for each of the electronic games in the sample set, respectively;

and the characteristic training unit is used for respectively training the initial vector of each electronic game to obtain the characteristic vector of each electronic game.

In an exemplary embodiment of the present invention, the feature training unit trains the initial vector of each of the electronic games based on a probability regression function, respectively.

In an exemplary embodiment of the present invention, the feature training unit includes:

the training target selection unit is used for respectively selecting each sample as a current sample and respectively selecting each electronic game in the current sample as a current electronic game;

the first sample training unit is used for combining the current electronic game and a positive sample element into a first sample; for the first example, with a label as 1 as a target, performing first training on the current electronic game and the initial vector of the positive sample element through the probability regression function;

In an exemplary embodiment of the present invention, the feature training unit further includes:

the second example training unit is used for combining the current electronic game and a negative example element into a second example; for the second example, taking a label as 0 as a target, and performing second training on the current electronic game and the initial vector of the negative sample element through the probability regression function;

In an exemplary embodiment of the invention, the apparatus further comprises:

the game clustering module is used for clustering all the electronic games according to the feature vectors of all the electronic games;

the first evaluation module is used for judging whether the characteristic vector of each electronic game reaches a target or not according to the clustering result; and when the feature vector of each electronic game is judged not to reach the target, training according to the sample set again to obtain the feature vector of the electronic game in each electronic game group.

In an exemplary embodiment of the invention, the apparatus further comprises:

the behavior acquisition module is used for acquiring the operation behavior of the target object on the target electronic game;

the second evaluation module is used for judging whether the characteristic vector of each electronic game reaches a target or not according to the operation behavior of the target object; and when the feature vector of each electronic game is judged not to reach the target, training according to the sample set again to obtain the feature vector of the electronic game in each electronic game group.

In an exemplary embodiment of the present invention, the recommendation judging module determines the degree of preference of the target object for the electronic game according to the following steps:

In an exemplary embodiment of the present invention, the recommendation judging module determines the feature vector of the target object according to the following steps:

According to a third aspect of the present invention, there is provided an electronic apparatus comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform the method of any one of the above via execution of the executable instructions.

According to a fourth aspect of the invention, there is provided a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of any one of the above.

Exemplary embodiments of the present invention may have the following advantageous effects:

in the electronic game recommendation method provided by the exemplary embodiment of the present invention, the electronic game groups associated with the respective users are respectively used as samples, and the feature vectors of the respective electronic games are obtained by training according to the samples, so that whether the candidate electronic game is used as the recommended target electronic game can be determined according to the similarity between the feature vector of the candidate electronic game and the feature vector of the electronic game contacted by the user. On one hand, the training is based on the electronic game groups associated with the users, and the electronic game feature vectors obtained by training can naturally reflect whether the electronic games can belong to the same electronic game group or not to a great extent; the essence of the electronic game recommendation is how to more accurately add new electronic games capable of being attributed to the original electronic game group of the user; therefore, the electronic game feature vector obtained based on the method of the invention not only can be directly used for recommending the electronic game, but also can realize more accurate electronic game recommendation, thereby realizing better electronic game recommendation effect, leading the recommendation resource to be more fully utilized and simultaneously improving the experience of the user to a great extent. On the other hand, in the electronic game recommendation method, the acquired data are only the electronic game groups associated with the users, and the demand degree for other data is not particularly high, so that the realization cost and the realization difficulty are reduced, and the method has stronger practicability.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention. It is obvious that the drawings in the following description are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.

FIG. 1 is a diagram illustrating an exemplary system architecture of an electronic game recommendation method and apparatus to which embodiments of the present invention may be applied;

FIG. 2 illustrates a schematic structural diagram of a computer system suitable for use with the electronic device to implement an embodiment of the invention;

FIG. 3 schematically illustrates a flow diagram of a method of electronic game recommendation in accordance with one embodiment of the present invention;

FIG. 4 schematically shows a flow chart of the steps of training a feature vector of an electronic game in accordance with an embodiment of the invention;

FIG. 5 schematically shows another flow chart of the steps of training feature vectors of an electronic game in accordance with an embodiment of the invention;

FIG. 6 schematically shows a flow chart of the steps of selecting a current electronic game in accordance with one embodiment of the present invention;

FIG. 7 schematically shows a further flowchart of the step of training a feature vector of an electronic game in accordance with an embodiment of the invention;

FIG. 8 schematically shows a flow chart of the steps of optimizing feature vectors of an electronic game in accordance with an embodiment of the invention;

FIG. 9 schematically shows another flow chart of the steps of optimizing feature vectors for an electronic game in accordance with an embodiment of the invention;

FIG. 10 schematically illustrates a block diagram of an electronic game recommendation device in accordance with one embodiment of the present invention;

FIG. 11 schematically shows a block diagram of a feature determination module according to one embodiment of the invention;

FIG. 12 schematically shows a block diagram of a feature training unit according to an embodiment of the present invention;

FIG. 13 schematically shows another block diagram of an electronic game recommendation device in accordance with one embodiment of the present invention.

Detailed Description

Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in many different forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to provide a thorough understanding of embodiments of the invention. One skilled in the relevant art will recognize, however, that the invention may be practiced without one or more of the specific details, or with other methods, components, devices, steps, and so forth. In other instances, well-known technical solutions have not been shown or described in detail to avoid obscuring aspects of the invention.

Furthermore, the drawings are merely schematic illustrations of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus their repetitive description will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor devices and/or microcontroller devices.

Fig. 1 is a schematic diagram illustrating a system architecture of an exemplary application environment to which an electronic game recommendation method and apparatus according to an embodiment of the present invention may be applied.

As shown in fig. 1, the system architecture 100 may include one or more of terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the terminal devices 101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few. The terminal devices 101, 102, 103 may be various electronic devices having a display screen, including but not limited to desktop computers, portable computers, smart phones, tablet computers, and the like. It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation. For example, server 105 may be a server cluster comprised of multiple servers, or the like.

The electronic game recommendation method provided by the embodiment of the invention is generally executed by the server 105, and accordingly, the electronic game recommendation device is generally arranged in the server 105. However, it is easily understood by those skilled in the art that the electronic game recommendation method provided in the embodiment of the present invention may also be executed by the terminal devices 101, 102, and 103, and accordingly, the electronic game recommendation apparatus may also be disposed in the terminal devices 101, 102, and 103, which is not particularly limited in this exemplary embodiment.

FIG. 2 illustrates a schematic structural diagram of a computer system suitable for use with the electronic device to implement an embodiment of the invention.

It should be noted that the computer system 200 of the electronic device shown in fig. 2 is only an example, and should not bring any limitation to the functions and the scope of the application of the embodiment of the present invention.

As shown in fig. 2, the computer system 200 includes a Central Processing Unit (CPU)201 that can perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)202 or a program loaded from a storage section 208 into a Random Access Memory (RAM) 203. In the RAM 203, various programs and data necessary for system operation are also stored. The CPU201, ROM 202, and RAM 203 are connected to each other via a bus 204. An input/output (I/O) interface 205 is also connected to bus 204.

The following components are connected to the I/O interface 205: an input portion 206 including a keyboard, a mouse, and the like; an output section 207 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage section 208 including a hard disk and the like; and a communication section 209 including a network interface card such as a LAN card, a modem, or the like. The communication section 209 performs communication processing via a network such as the internet. A drive 210 is also connected to the I/O interface 205 as needed. A removable medium 211 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 210 as necessary, so that a computer program read out therefrom is mounted into the storage section 208 as necessary.

In particular, according to an embodiment of the present invention, the processes described below with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the invention include a computer program product comprising a computer program embodied on a computer-readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 209 and/or installed from the removable medium 211. The computer program, when executed by a Central Processing Unit (CPU)201, performs various functions defined in the methods and apparatus of the present application.

It should be noted that the computer readable medium shown in the present invention can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present invention, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present invention, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The units described in the embodiments of the present invention may be implemented by software, or may be implemented by hardware, and the described units may also be disposed in a processor. Wherein the names of the elements do not in some way constitute a limitation on the elements themselves.

As another aspect, the present application also provides a computer-readable medium, which may be contained in the electronic device described in the above embodiments; or may exist separately without being assembled into the electronic device. The computer readable medium carries one or more programs which, when executed by an electronic device, cause the electronic device to implement the method as described in the embodiments below. For example, the electronic device may implement the steps shown in fig. 3 to 9, and the like.

The technical scheme of the embodiment of the invention is explained in detail as follows:

in the related art, in order to accurately recommend an electronic game to a user, one scheme is as follows: generating a feature vector of the electronic game based on the category of the electronic game, and further judging whether the candidate electronic game is taken as a recommended electronic game according to the similarity between the feature vector of the candidate electronic game and the feature vector of the user; in the scheme, the information contained in the feature vector can only distinguish the categories of the electronic games, and when the similarity between the electronic games is measured, only two values of the same category and the non-same category can be output, so that the contained information quantity is less, and the judgment mode is rough. The other scheme is as follows: generating a characteristic vector through a text theme model and other modes based on the text data of the electronic game, and further judging whether the candidate electronic game is taken as a recommended electronic game according to the similarity between the characteristic vector of the candidate electronic game and the characteristic vector of the user; in the scheme, the feature vectors can be well represented in the dimensions of the electronic game category, the theme and the like, but the contained information is still more unilateral, and all factors influencing the user experience, such as game heat, game quality and the like, cannot be covered, so that the problem of poor recommendation accuracy still exists. In addition, the electronic game feature vectors generated in the prior art are generally difficult to be directly used for electronic game recommendation; for example, in the prior art, the feature vector of the electronic game generated based on the category mode of the electronic game essentially reflects the category attribute of the game, so that the feature vector cannot be directly used for recommending the electronic game, and needs to determine the category preference of the user and perform other related operations.

In view of the above, the present exemplary embodiment first provides an electronic game recommendation method. The electronic game recommendation method may be applied to the server 105, and may also be applied to one or more of the terminal devices 101, 102, and 103, which is not particularly limited in this exemplary embodiment. Referring to fig. 3, the electronic game recommendation method may include the steps of:

s310, respectively taking the electronic game groups associated with the sampling objects as samples to form a sample set; wherein the number of electronic games in the set of electronic games is greater than 1;

s320, training according to the sample set to obtain a feature vector of the electronic game in each electronic game group;

s330, determining a feature vector of a target object, and calculating the similarity between the feature vector of the target object and the feature vector of the electronic game; determining the preference degree of the target object to the electronic game according to the similarity, and if the preference degree of the target object to the electronic game is greater than a preset threshold value, determining that the electronic game is the target electronic game; and

and step S340, recommending the determined target electronic game to the target object.

In the electronic game recommendation method provided in this exemplary embodiment, the electronic game groups associated with the respective users are respectively used as samples, and the feature vectors of the respective electronic games are obtained by training according to the samples, and further, whether the candidate electronic game is used as the recommended target electronic game can be determined according to the similarity between the feature vector of the candidate electronic game and the feature vector of the electronic game contacted by the user. On one hand, the training is based on the electronic game groups associated with the users, and the electronic game feature vectors obtained by training can naturally reflect whether the electronic games can belong to the same electronic game group or not to a great extent; the essence of the electronic game recommendation is how to more accurately add new electronic games capable of being attributed to the original electronic game group of the user; therefore, the electronic game feature vector obtained based on the method of the invention not only can be directly used for recommending the electronic game, but also can realize more accurate electronic game recommendation, thereby realizing better electronic game recommendation effect, leading the recommendation resource to be more fully utilized and simultaneously improving the experience of the user to a great extent. On the other hand, in the electronic game recommendation method, the acquired data are only the electronic game groups associated with the users, and the demand degree for other data is not particularly high, so that the realization cost and the realization difficulty are reduced, and the method has stronger practicability.

The above steps of the present exemplary embodiment will be described in more detail below.

In step S310, the electronic game groups associated with the respective sampling objects are each set as samples to constitute a sample set.

In this example embodiment, the sample object may comprise part or all of a user of various electronic games. For each sample object, there may be one or more electronic games associated with it; all the electronic games associated with one sampling object are the electronic game group associated with the sampling object. The association described in this exemplary embodiment may include purchase of an electronic game, download of the electronic game, account registration of the electronic game, creation of a virtual character in the electronic game, an active time in the electronic game exceeding a preset time period, consumption exceeding a preset amount in the electronic game, and the like, and this is not limited in this exemplary embodiment. The specific sampling manner may be an electronic game group associated with each sampling object acquired through data such as user information and historical game records of each electronic game platform, or an electronic game group associated with each sampling object acquired through other manners such as questionnaire survey, which is not particularly limited in this exemplary embodiment.

In addition, the electronic games that the user is interested in may change over time. Therefore, in the present exemplary embodiment, the time interval for acquiring data may be defined; for example, an electronic game in which the sampling object has been contacted for a preset time may be acquired. For example, an electronic game in which the sampling object is acquired to have logged in the last three months, an electronic game in which the cumulative active time of the sampling object in the last year exceeds 30 hours, an electronic game in which the sampling object has a recharge record in the last 2 years, an electronic game in which the sampling object has been registered in the last three years, or the like, an electronic game in which the sampling object has created a virtual character in the last three years, or the like. The preset time may be specifically set by a person skilled in the art according to specific needs, and is not particularly limited in the exemplary embodiment. Of course, in other exemplary embodiments of the present invention, the time interval for acquiring data may not be limited, for example, all electronic games that are logged in by the user in history, all electronic games that all active time of the sampling object in history exceeds the preset time duration, all electronic games that have a recharging record of the sampling object, all electronic games that have been registered in history of the sampling object, and the like, all electronic games that have virtual characters created by the sampling object, and the like are acquired, which also belongs to the protection scope of the present invention.

If the electronic game group associated with one sampling object only comprises one electronic game, the electronic game group does not exist, and therefore the electronic game cannot be used or has no reference value in subsequent training; if too many video games are included in the video game set associated with one sample object, the correlation between the video games is reduced, and thus the reference value is not high as well. Based on this, in the present exemplary embodiment, the collected data may be filtered to obtain the electronic game groups with the number of electronic games greater than 1 as samples; meanwhile, samples with the number of the electronic games larger than a preset threshold value can be eliminated. The preset threshold may be specifically set by those skilled in the art according to specific needs, such as 5, 10, etc., which is not particularly limited in the exemplary embodiment. In addition, the electronic game groups comprising too many electronic games can be sorted according to a certain rule, and only the electronic games which are close to each other are reserved; for example, an electronic game ranked 10 top of the average day active time is reserved, and so on.

The resulting sample may be, for example, sample 1: [ electronic Game G₁Electronic game G₂Electronic game G₃]Sample 2: [ electronic Game G₁Electronic game G₆]Sample 3: [ electronic Game G₂Electronic game G₄Electron, electronGame G₅Electronic game G₆]Sample 4: [ electronic Game G₁Electronic game G₂Electronic game G₃Electronic game G₅Electronic game G₆]And the like; the set of these samples is the sample set in this example embodiment. In each sample, the sequence of the electronic game may be a random arrangement or a specific regular arrangement, which is not particularly limited in the exemplary embodiment. In addition, in the present exemplary embodiment, each electronic game may be identified by an electronic game name or other unique identifier of the electronic game, which is also not particularly limited in the present exemplary embodiment.

And S320, training according to the sample set to obtain the feature vectors of the electronic games in the electronic game groups. In the present exemplary embodiment, an example will be described in which feature vectors of respective video games are obtained by training a neural network model. However, it is easily understood by those skilled in the art that in other exemplary embodiments of the present invention, the feature vector of each electronic game may be obtained in more ways, such as by training a random forest model, and the like, and the present invention also belongs to the protection scope of the present invention.

For example, referring to fig. 4, in the present exemplary embodiment, step S320 may include step S410 and step S420. Wherein:

in step S410, an initial vector is determined for each of the electronic games in the sample set, respectively.

For example, for each video game G_iAn N-dimensional vector W may be initialized_i＝(x_i1,x_i2,…,x_iN) And randomly generating a value of each dimension of the vector to obtain an initial vector, for example, the value of each dimension of the initial vector can be [ -1,1]In the meantime. The value of N may be determined according to the number of the electronic games, for example, the larger the number of the electronic games is, the larger the corresponding value of N is; the smaller the number of electronic games, the smaller the value of the corresponding N. To say thatIt is clear that each dimension takes on a value of [ -1,1 [)]In other exemplary embodiments of the present invention, the value of each dimension may also be [0,1 ]]And the like, which are not particularly limited in this exemplary embodiment.

In step S420, the initial vector of each electronic game is trained to obtain a feature vector of each electronic game.

Taking the feature vectors of each electronic game obtained through training of the neural network model as an example, the neural network model generally includes an input layer, a mapping layer, and an output layer. In the present exemplary embodiment, for each of the electronic games G_iThe input of the input layer of the neural network model may be G_iAnd G_iElectronic game G on the same sample_mThe output of the output layer of the neural network model may be an electronic game G_iAnd an electronic game G_mProbabilities of being in the same electronic game group; the mapping layer of the neural network model can map the input layer to the output layer through a mapping function, wherein the mapping function comprises two parameters, and each parameter comprises N dimensions; after the neural network model is trained, two N-dimensional vectors corresponding to the electronic game G can be obtained_iCharacteristic vector W of_iAnd an electronic game G_mCharacteristic vector W of_m. Specifically, the method comprises the following steps:

in this exemplary embodiment, since the finally obtained feature vector of the electronic game needs to reflect the probability that the feature vector and another electronic game can be located in the same electronic game group to some extent, the mapping function may be a probability regression function; however, the exemplary embodiment is not limited thereto, and other types of mapping functions may be used, which also falls within the scope of the present invention. On this basis, referring to fig. 5, training the initial vector of each of the electronic games separately may include steps S510 to S530. Wherein:

in step S510, each of the samples is selected as a current sample, and each of the electronic games in the current sample is selected as a current electronic game. For example, in the present exemplary embodiment, the steps may include steps S610 to S650 shown in fig. 6. Wherein:

in step S610, the sample set is acquired.

In step S620, it is determined whether there are unselected samples in the sample set; if there are unselected samples, go to step S630, and if there are no unselected samples, end. In addition, in the present exemplary embodiment, the selected samples in the sample set may be labeled or the like, so as to determine whether there are unselected samples in the sample set.

In step S630, a non-selected sample is selected as the current sample. For example, the unselected samples may be sequentially selected as the current sample, or the unselected samples may be randomly selected or selected according to other rules as the current sample, which is not particularly limited in this exemplary embodiment.

In step S640, it is determined whether there is an unselected electronic game in the current sample; if there is an unselected video game, the process goes to step S660, and if there is no unselected video game, the process ends. In addition, in the present exemplary embodiment, the electronic games selected in the current sample set may be marked, so as to determine whether there are unselected electronic games in the current sample set.

In step S650, an unselected video game is selected as the current video game. For example, the unselected electronic games may be sequentially selected as the current electronic game, or the unselected electronic games may be randomly selected or selected according to other rules as the current electronic game, which is not limited in this exemplary embodiment.

In step S520, the current electronic game and a positive sample element are combined into a first sample.

In this example embodiment, the positive sample element may be an initial vector of an electronic game other than the current electronic game in the current sample. For example, with the current sample including: [ electronic Game G_l：W_l＝(x_j1,x_j2,…,x_jN) Electronic game G_i：W_i＝(x_i1,x_i2,…,x_iN) Electronic game G_j：W_j＝(x_j1,x_j2,…,x_jN) Electronic game G_m：W_m＝(x_m1,x_m2,…,x_mN)]The current electronic game is an electronic game G_iFor example, the remaining video game G may be selected from the current sample_l、G_j、G_mRandomly selecting one as a positive sample element, e.g. G_m. Then the electronic game G_iAnd an electronic game G_mConstitute a first example.

In step S530, for the first example, with the label as 1 as a target, the initial vector of the current electronic game and the positive sample element is subjected to a first training through the probability regression function.

In the first example described above, two electronic games G_iAnd G_mHave appeared in the same electronic game group, and thus electronic game G_iAnd an electronic game G_mThe probability of being located in the same electronic game group, i.e. the value of the tag, may be 1. In this example embodiment, the probability regression function may be as follows:

wherein, P (W)_m|W_i) Presentation video game G_mAnd an electronic game G_iProbabilities of being located in the same electronic game group. W_i·W_mThe larger the value of (A), the larger P (W)_m|W_i) The closer to 1.

In the specific training process, based on the maximum likelihood estimation method, the current electronic game and the initial vector of the positive sample element are simultaneously subjected to first training through algorithms such as random gradient descent and the like, namely two parameters W of the probability regression function are simultaneously optimized_mAnd W_i：

Wherein,the learning rate is usually determined from an empirical value, and is, for example, 0.1, 0.05, or the like.

Upon reaching a preset number of iterations, or P (W)_m|W_i) If the specified threshold has been met, the operation may be stopped; at this time, W_mAnd W_iCan be used as an electronic game G_iAnd an electronic game G_mThe feature vector of (2).

Further, in order to improve the training efficiency, reduce the training time, and improve the reliability of the training result to a certain extent, in the present exemplary embodiment, a negative sample element may be further introduced to perform the second training. For example, referring to fig. 7, training the initial vector of each of the electronic games may further include step S540 and step S550. Wherein:

in step S540, the current video game and a negative sample element are combined into a second sample.

In this example embodiment, the negative sample element is an electronic game in a sample other than the current sample in the sample set. For example, with the current sample including: [ electronic Game G_l：W_l＝(x_l1,x_l2,…,x_lN) Electronic game G_i：W_i＝(x_i1,x_i2,…,x_iN) Electronic game G_j：W_j＝(x_j1,x_j2,…,x_jN) Electronic game G_m：W_m＝(x_m1,x_m2,…,x_mN)]The current electronic game is an electronic game G_iFor example, an electronic game may be selected from samples other than the current sample, such as electronic game G_kAs a negative sample element, the video game G_iAnd an electronic game G_kConstitute a second example.

It should be noted that, for the selection of the negative sample element, an electronic game may be randomly selected from samples other than the current sample; an electronic game can be selected from samples except the current sample in a weighted sampling mode, so that the probability that the electronic game with higher occurrence frequency is selected as a negative sample element is higher, and the accuracy of the training result is improved; this is not particularly limited in the present exemplary embodiment. In addition, if the selected negative example element is an electronic game belonging to the current example, such as the electronic game G_l、G_i、G_jAnd G_mThen the selection can be made again.

In step S550, for the second example, with the label as 0 as the target, performing a second training on the current electronic game and the initial vector of the negative sample element through the probability regression function.

In the second example described above, two electronic games G_iAnd G_kNot present in the same electronic game group, and thus electronic game G_iAnd an electronic game G_kThe probability of being in the same electronic game group, i.e. the value of the tag, may be 0. In this example embodiment, the probability regression function may be as follows:

wherein, P (W)_k|W_i) Presentation video game G_iAnd an electronic game G_kProbabilities of being located in the same electronic game group. W_i·W_kThe smaller the value of (A), the smaller P (W)_k|W_i) The closer to 0.

In the specific training process, based on the maximum likelihood estimation method, the current electronic game and the initial vector of the negative sample element are simultaneously subjected to second training through algorithms such as random gradient descent and the like, namely two parameters W of the probability regression function are simultaneously optimized_kAnd W_i：

Upon reaching a preset number of iterations, or P (W)_k|W_i) If the specified threshold has been met, the operation may be stopped; at this time, W_kAnd W_iCan be used as an electronic game G_kAnd an electronic game G_iThe feature vector of (2).

The first training and the second training can be performed simultaneously in the same neural network model, i.e. in two ways, simultaneously for W_iOptimizing; therefore, the training efficiency can be improved, the training time is reduced, and the reliability of the training result is improved to a certain degree.

Through the steps S510 to S550, the feature vector of the current electronic game can be obtained; the above steps S510 to S550 are performed for each electronic game in each sample, and feature vectors of all electronic games can be obtained.

It should be noted that, as is readily understood by those skilled in the art, in other exemplary embodiments of the present invention, the feature vector of each electronic game may also be obtained through the above-mentioned training of the random forest model; even in the above neural network model, those skilled in the art can select other regression functions, select other parameter optimization algorithms such as newton method, or design other training processes, etc. according to the needs; all of which are also within the scope of the present invention.

In step S330, determining a feature vector of a target object, and calculating a similarity between the feature vector of the target object and a feature vector of the electronic game; and determining the preference degree of the target object to the electronic game according to the similarity, and if the preference degree of the target object to the electronic game is greater than a preset threshold value, determining that the electronic game is the target electronic game.

In the present exemplary embodiment, the target object is a user to whom the electronic game is to be recommended. Determining the characteristic vector of the target object in different modes according to different types of target objects; meanwhile, according to different requirements, different modes can be selected to determine the feature vector of the target object. For example:

for example, for a target object which is already associated with an electronic game, feature vectors of the electronic games associated with the target object may be respectively used as feature vectors of the target object, then, similarity between each feature vector of the target object and a feature vector of one electronic game is respectively calculated, and whether the electronic game is the target electronic game is determined according to each similarity. For example:

if the target object is associated with an electronic game G₁:W₁，G₂:W₂，…，G_m:W_m](ii) a Preparing a candidate video game to be recommended to a target object as video game G_n:W_nThen the feature vector W can be transformed₁，W₂，…，W_mRespectively as the feature vectors of the target object; accordingly, the candidate electronic games G are respectively calculated_nAnd the feature vector Wn of the target object₁，W₂，…，W_mSimilarity between them Y₁，Y₂，…，Y_m(ii) a Finally, according to the similarity data, the preference degree of the target object to the candidate electronic game is determined, and then the candidate electronic game G is determined_nWhether it is a target video game. For example:

can utilize the similarity Y₁，Y₂，…，Y_mThe maximum value in (1) represents the preference degree Lm of the target object for the candidate electronic game; i.e. Lm Max (Y)_i). The similarity Y can also be used₁，Y₂，…，Y_mRepresents the degree of preference Lm of the target object for the candidate electronic game; namely, it isThe similarity Y can also be used₁，Y₂，…，Y_mRepresents the preference degree Lm of the target object for the candidate electronic game; namely, it isWherein w_iFor each degree of similarity Y_iThe specific value of the weight (c) can be specifically set according to actual conditions, for example, Y can be_iThe larger the corresponding weight w_iThe larger, etc. Of course, in other exemplary embodiments of the present invention, the degree of preference of the target object for the candidate video game may also be determined by using the similarity data in other manners, such as taking a median value, which is not particularly limited in this exemplary embodiment.

After determining the preference degree of the target object for the candidate electronic game, it may be determined whether the candidate electronic game is the target electronic game accordingly; for example, if the preference degree of the target object for the candidate video game is greater than a preset threshold, the candidate video game is determined to be the target video game. The preset threshold may be specifically set by those skilled in the art according to actual needs, and is not particularly limited in the present exemplary embodiment.

For example, for a target object that is already associated with an electronic game, feature vectors of all the electronic games associated with the target object may be calculated to obtain a feature vector of the target object. For example:

for example, a target object is associated with an electronic game G₁:W₁，G₂:W₂，…，G_m:W_m](ii) a Preparing a candidate video game to be recommended to a target object as video game G_n:W_nThen can be to W₁，W₂，…，W_mCalculating to obtain the characteristic vector W of the target object_x. For example, a feature vector W of an electronic game is associated with a target object₁，W₂，…，W_mCarrying out operation such as averaging and the like to obtain a characteristic vector W of the target object_x. Finally, candidate video games G are calculated_nCharacteristic vector W of_nFeature vector W with target object_xSimilarity between them Y_xAnd further according to the similarity Y_xTo determine candidate electronic games G_nWhether it is a target video game. For example, if the similarity Y_xIf the value of the game is larger than a preset threshold value, the candidate electronic game is judged to be the target electronic game. The preset threshold may be specifically set by those skilled in the art according to actual needs, and is not particularly limited in the present exemplary embodiment.

As another example, if a target object is not associated with any video game, a related object having a preset relationship with the target object may be first determined. In this exemplary embodiment, the preset relationship may be, for example: in social applications, a friend relationship exists between a target object and a related object; the target object and the related objects are members of one or more virtual communities; the user images of the target object and the related object have a higher similarity, and the like, which is not particularly limited in this exemplary embodiment. After determining the relevant object to the target object, then a feature vector for the target object may be determined from the feature vectors of all of the electronic games associated with the relevant object. For example:

if the object associated with the target object is associated with an electronic game G₁:W₁，G₂:W₂，…，G_m:W_m](ii) a Preparing a candidate video game to be recommended to a target object as video game G_n:W_nThen W can be converted to₁，W₂，…，W_mRespectively as the feature vectors of the target object; accordingly, the candidate electronic games G are respectively calculated_nAnd the feature vector Wn of the target object₁，W₂，…，W_mSimilarity between them Y₁，Y₂，…，Y_m(ii) a Finally, according to the similarity data, the preference degree of the target object to the candidate electronic game is determined, and then the candidate electronic game G is determined_nWhether it is a target video game. Alternatively, W may be the same as W₁，W₂，…，W_mCalculating to obtain the characteristic vector W of the target object_xRecalculating candidate video games G_nCharacteristic vector W of_nSimilarity Y with feature vector Wx of target object_xAnd further according to the similarity Y_xTo determine candidate electronic games G_nWhether it is a target electronic game, etc.; for example, if the similarity Y_xIf the value of the game is larger than a preset threshold value, the candidate electronic game is judged to be the target electronic game. The preset threshold may be specifically set by those skilled in the art according to actual needs, and is not particularly limited in the present exemplary embodiment.

Of course, besides the above examples, those skilled in the art may determine the feature vector of the target object in other manners, and determine whether the candidate video game is the target video game according to the feature vector of the target object and the feature vector of the candidate video game in other manners, which all belong to the protection scope of the present invention. In addition, the feature vector of the target object may be further fine-tuned according to other information of the target object, for example, information such as age, gender, credit level, consumption level, and the like, which is not particularly limited in this exemplary embodiment.

In the present exemplary embodiment, the specific manner of recommending the target electronic game to the target object is not particularly limited; for example, a time period during which the target object frequently logs in the game may be recorded, and the target electronic game may be recommended to the target object in the form of a pop-up window during the time period; or setting an electronic game recommendation function key in an interface of the electronic game platform, and recommending the target electronic game after the user clicks the function key; or recommending the target electronic game in an advertisement position in application software used by the target object; or, actively pushing the related notice of the target electronic game to the terminal device of the target object, and the like.

In addition, in the present exemplary embodiment, the training result of the feature vector of each electronic game may be evaluated, and the feature vector of each electronic game may be optimized according to the evaluation result.

For example, referring to fig. 8, the training result of the feature vector of each video game can be evaluated and optimized through steps S810 and S820. Specifically, the method comprises the following steps:

in step S810, all the electronic games may be clustered according to the feature vectors of the electronic games. For example, according to the feature vector of each electronic game, all the electronic games are clustered through a clustering algorithm such as a k-means algorithm, a k-means algorithm or a clara algorithm; taking the k-means algorithm as an example, the clustering process may include: firstly, randomly selecting a preset number of electronic games as initial clustering centers, secondly, calculating the distance between the rest electronic games and the initial clustering centers in a vector space according to the characteristic vectors, then distributing each electronic game to the initial clustering center closest to the electronic game, finally, recalculating the initial clustering centers according to the distribution results, and repeating the process until the end condition of clustering is met.

In step S820, the training result of the feature vector of each electronic game may be evaluated according to the result of the clustering. For example, after the clustering is completed, the clustering result may be first visualized, for example, by presenting the clustering result in a graph or the like, so as to facilitate evaluation of the evaluation value. Criteria for evaluation may include, for example: similar play, whether electronic games with the same game elements are located in the same clustering result, whether electronic games with larger user scale are clustering centers, and the like.

In step S830, if it is determined that the result of the clustering satisfies the evaluation criterion, the feature vector of each of the electronic games may be considered to reach the target. On the contrary, if the feature vector of each electronic game does not reach the target, the feature vector of the electronic game in each electronic game group needs to be obtained by training again according to the sample set; for example, the dimensions of the feature vectors are rearranged, more samples are obtained for training, etc.

For another example, referring to fig. 9, the training result of the feature vector of each video game can be evaluated and optimized through steps S910 and S920. Specifically, the method comprises the following steps:

in step S910, the operation behavior of the target object with respect to the recommended target electronic game is acquired. For example, click behavior, download behavior, account registration behavior, subsequent game behavior, uninstall behavior, and the like of the target object for the target electronic game are acquired. The operation behavior of the target object for the recommended target electronic game may be obtained through an application software downloading platform, or may be obtained through other manners such as obtaining through an electronic game server, which is not particularly limited in this exemplary embodiment.

In step S920, a training result of the feature vector of each electronic game is evaluated according to the operation behavior of the target object. For example, if the operation behavior of the target object on the target electronic game is detected, the electronic game recommendation may be considered to be valid, and meanwhile, the effect of the electronic game recommendation may be evaluated in more detail according to the game time of the user in the target electronic game, for example, if the active duration of the target object in the target electronic game exceeds a certain duration, the effect of the electronic game recommendation may be considered to be better; on the contrary, if it is detected that the user unloads the target electronic game in a short time or does not acquire any of the above-described operation behaviors of the user, it may be considered that the effect of the electronic game recommendation is not good and does not satisfy the evaluation criterion.

In step S930, if it is determined that the effect recommended by the electronic game does not satisfy the evaluation criterion, determining that the feature vector of the electronic game does not reach the target; then, training according to the sample set again to obtain the feature vectors of the electronic games in each electronic game group; for example, the dimensions of the feature vectors are rearranged, more samples are obtained for training, etc.

Of course, in other exemplary embodiments of the present invention, the training result of the feature vector of each electronic game may be evaluated in other manners, which is not limited in this exemplary embodiment.

It should be noted that although the steps of the methods of the present invention are depicted in the drawings in a particular order, this does not require or imply that the steps must be performed in this particular order, or that all of the depicted steps must be performed, to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step execution, and/or one step broken down into multiple step executions, etc.

Further, in the present exemplary embodiment, an electronic game recommendation apparatus is also provided. The electronic game recommendation device can be applied to a server or terminal equipment. Referring to fig. 10, the electronic game recommendation apparatus 1000 may include a sample set construction module 1010, a feature determination module 1020, a recommendation judgment module 1030, and a game recommendation module 1040. Wherein:

the sample set constructing module 1010 may be configured to form a sample set by taking the electronic game groups associated with the respective sampling objects as samples; wherein the number of electronic games in the set of electronic games is greater than 1; the feature determining module 1020 may be configured to obtain feature vectors of the electronic games in each of the electronic game groups according to the sample set training; the recommendation judging module 1030 may be configured to determine a feature vector of a target object, and calculate a similarity between the feature vector of the target object and a feature vector of the electronic game; determining the preference degree of the target object to the electronic game according to the similarity, and if the preference degree of the target object to the electronic game is greater than a preset threshold value, determining that the electronic game is the target electronic game; and, the game recommending module 1040 may be configured to recommend the determined target electronic game to the target object.

Referring to fig. 11, in an exemplary embodiment of the invention, the feature determination module 1020 includes:

a vector initialization unit 1110, configured to determine initial vectors for the electronic games in the sample set respectively;

the feature training unit 1120 is configured to train the initial vector of each electronic game to obtain a feature vector of each electronic game.

In an exemplary embodiment of the present invention, the feature training unit 1120 trains the initial vector of each of the electronic games based on a probability regression function.

Referring to fig. 12, in an exemplary embodiment of the present invention, the feature training unit 1120 includes:

a training target selecting unit 1210, configured to select each sample as a current sample, and select each electronic game in the current sample as a current electronic game;

a first example training unit 1220, configured to combine the current electronic game and a positive sample element into a first example; for the first example, with a label as 1 as a target, performing first training on the current electronic game and the initial vector of the positive sample element through the probability regression function;

With continued reference to FIG. 12, in an exemplary embodiment of the invention, the feature training unit 1120 further comprises:

a second example training unit 1230, configured to combine the current electronic game and a negative example element into a second example; for the second example, taking a label as 0 as a target, and performing second training on the current electronic game and the initial vector of the negative sample element through the probability regression function;

Referring to fig. 13, in an exemplary embodiment of the invention, the apparatus further comprises:

a game clustering module 1310, configured to cluster all the electronic games according to the feature vectors of the electronic games;

a first evaluation module 1320, configured to determine whether the feature vector of each electronic game reaches a target according to the clustering result; and when the feature vector of each electronic game is judged not to reach the target, training according to the sample set again to obtain the feature vector of the electronic game in each electronic game group.

In an exemplary embodiment of the invention, the apparatus further comprises:

a behavior obtaining module 1330, configured to obtain an operation behavior of the target object on the target electronic game;

the second evaluation module 1340 is configured to determine whether the feature vector of each electronic game reaches a target according to the operation behavior of the target object; and when the feature vector of each electronic game is judged not to reach the target, training according to the sample set again to obtain the feature vector of the electronic game in each electronic game group.

In an exemplary embodiment of the present invention, the recommendation judging module 1030 determines the preference degree of the target object for the video game according to the following steps:

In an exemplary embodiment of the present invention, the recommendation determining module 1030 determines the feature vector of the target object according to the following steps:

The specific details of each module in the electronic game recommendation device have been described in detail in the corresponding electronic game recommendation method, and therefore are not described herein again.

It should be noted that although in the above detailed description several modules or units of the device for action execution are mentioned, such a division is not mandatory. Indeed, the features and functionality of two or more modules or units described above may be embodied in one module or unit, according to embodiments of the invention. Conversely, the features and functions of one module or unit described above may be further divided into embodiments by a plurality of modules or units.

Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.

It will be understood that the invention is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the invention is limited only by the appended claims.

Claims

1. An electronic game recommendation method, comprising:

recommending the determined target electronic game to the target object.

2. The video game recommendation method of claim 1, wherein the sample object associated video game set comprises one or more of: the electronic game which is contacted by the sampling object in a preset time, the historical registration electronic game of the sampling object, the electronic game which has the active time of the sampling object meeting a preset time length, and the electronic game which has the consumption record of the sampling object.

3. The method for recommending electronic games according to claim 1, wherein said training according to said sample set to obtain feature vectors of electronic games in each of said electronic game groups comprises:

4. The method of claim 3, wherein the training of the initial vector for each of the video games comprises:

5. The method of claim 4, wherein training an initial vector for each of the video games comprises:

6. The method of claim 5, wherein training an initial vector for each of the video games is performed separately, further comprising:

7. The method for recommending an electronic game according to any of claims 3 to 6, further comprising:

8. The method for recommending an electronic game according to any of claims 3 to 6, further comprising:

9. The electronic game recommendation method of claim 1, wherein determining a preference level of a target object for the electronic game based on each of the similarities comprises:

10. The video game recommendation method of claim 1, wherein determining a feature vector for a target object comprises:

computing all the feature vectors of the electronic game associated with the target object to obtain the feature vector of the target object, or