A game recommendation method based on big data analysis and a multi-game integration platform

Abstract

The application relates to a game recommendation method based on big data analysis and a multi-game integrated platform, and relates to the technical field of games, which comprises the following steps: S1: collecting historical games; S2: responding to the historical games to call historical playing time; S3: defining the historical game with the longest historical playing time as a marked historical game; S4: responding to the marked historical game to call historical screen parameters and a marked historical type; S5: responding to the marked historical type, the historical games and the historical screen parameters to obtain a recommended game, and outputting the recommended game to a preset display area. The application has the effect of improving the efficiency of game recommendation.

Description

Technical Field

[0001] This invention relates to the field of game technology, and in particular to a game recommendation method and a multi-game integration platform based on big data analysis. Background Technology

[0002] A game is an activity based on certain rules, which is usually entertaining, interactive, and goal-oriented.

[0003] Games include video games. When a user is playing a video game, and the user searches for similar games based on the content and gameplay of the currently played game, the game platform usually only recommends games with corresponding gameplay based on the player's playtime. Then, the user selects a game to play based on the description of the recommended game.

[0004] When users need to find games, the game platform recommends a large number of games based solely on the player's playtime, and these games often have multiple gameplay styles. Some of these games may not be suitable for the user's playstyle, requiring the user to manually search for them based on the game descriptions, resulting in low efficiency in game recommendations. Summary of the Invention

[0005] To improve the efficiency of game recommendation, this invention provides a game recommendation method based on big data analysis and a multi-game integration platform.

[0006] Firstly, the present invention provides a game recommendation method based on big data analysis, employing the following technical solution:

[0007] A game recommendation method based on big data analysis includes:

[0008] S1: Collect historical games;

[0009] S2: Respond to the historical game to retrieve the historical playtime;

[0010] S3: Define the historical game with the longest historical playtime as the marked historical game;

[0011] S4: Retrieve the history marking game to retrieve history screen parameters and history marking type;

[0012] S5: In response to the marked history type, the history game, and the history screen parameters, a recommended game is obtained, and the recommended game is output to a preset display area.

[0013] By adopting the above technical solution, the historical screen parameters and marked historical types are obtained by analyzing the historical playtime of historical games. The historical screen parameters, marked historical types, and historical games are then analyzed to obtain recommended games. This allows for the automatic recommendation of games with similar gameplay to users, reducing the time users spend searching and improving the efficiency of users searching for games.

[0014] Optionally, the method for obtaining the recommended games includes:

[0015] S51: Responding to the historical game and the marking game to obtain the marking hue range;

[0016] S52: In response to the stated tag history type, select similar games;

[0017] S53: Respond to the similar game to retrieve the baseline display hue;

[0018] S54: Select the recommended game by comparing the marked hue range with the baseline display hue.

[0019] By adopting the above technical solution, the range of marked color tones is obtained by analyzing historical games and marked games, and the baseline display color tone is obtained by analyzing the marked historical types. Recommended games are selected by the inclusion of the marked color tone range and the baseline display color tone. In this way, games with similar color tones and types to historical games can be recommended to users, making it convenient for users to play.

[0020] Optionally, the method for obtaining the marked hue range includes:

[0021] S5101: Capture the marking game screen of the marking game;

[0022] S5102: Responding to the marked game screen to generate a color tone type;

[0023] S5103: Detect game in response to the tone type, the history game, and the marked history game;

[0024] S5104: In response to the detected game and the marked game screen, generate a detection screen;

[0025] S5105: Respond to the detection screen to obtain the marked hue range.

[0026] By adopting the above technical solution, the detected game is obtained by comparing the marked game screen with historical games, and the marked color range is obtained by comparing the detected game with the marked game screen. In this way, the marked color range can be retrieved from each historical game, thereby improving the accuracy of game recommendations.

[0027] Optional, also includes:

[0028] S5111: In response to the detected game, obtain the detected game type;

[0029] S5112: Obtain the marked detection range by detecting the game type;

[0030] S5113: Responding to the detection screen and the marked detection range to obtain a detection box selection screen;

[0031] S5114: Respond to the detection frame selection screen to generate a marker feature color;

[0032] S5115: Responding to the marker feature color and the preset reference color to obtain the marker hue parameter;

[0033] S5116: In response to the marker hue parameter, generate the marker hue range.

[0034] Optionally, the method for obtaining the marker hue parameter includes:

[0035] S5121: Select a color from the marked colors that is inconsistent with the reference color as a neutral color;

[0036] S5122: Responding to the historical screen parameters and the neutral color to obtain the neutral hue parameter;

[0037] S5123: Update the marker tone parameter using the neutral tone parameter.

[0038] Optionally, the method for generating the marker feature color further includes:

[0039] S51141: Collect image detection information from the user;

[0040] S51142: Responding to the image detection information to generate an emotional change image;

[0041] S51143: The duration of emotional changes is retrieved by comparing the emotional change images with the historical playtime.

[0042] S51144: Collect the interaction commands of the detected game;

[0043] S51145: Retrieve the interaction duration in response to the interaction command and the historical play duration;

[0044] S51146: Responding to the interaction duration and the emotion change duration to generate an emotion detection screen, and using the emotion detection screen to generate the marker feature color.

[0045] By adopting the above technical solution, the system detects emotional changes in users while playing games, extracts marker colors from these emotional images, and further understands users' interests in the game's color scheme or gameplay. This allows the system to retrieve corresponding color tones for game recommendations, thereby improving the accuracy of game recommendations.

[0046] Optionally, the method for generating the emotional change images includes:

[0047] S511421: Responding to the image detection information and preset facial features to obtain facial image information;

[0048] S511422: Update the facial image information in response to the interaction duration and the emotion change duration;

[0049] S511423: Response to the facial image information to obtain detected emotion;

[0050] S511424: Select the emotion change image from the detection frame screen by detecting the emotion.

[0051] Optionally, the method for verifying detected emotions includes:

[0052] S51142201: Obtain an eye image in response to the facial image information and preset eye features;

[0053] S51142202: Obtain the eye orientation from the eye image;

[0054] S51142203: When the eye orientation is inconsistent with the preset reference orientation, the image to be viewed is retrieved in response to the eye orientation and the image detection information;

[0055] S51142204: When the selected image contains a preset viewing feature, a labeled emotion image is obtained in response to the viewing feature, the reference orientation, and the eye orientation;

[0056] S51142205: Update the facial image information in response to the labeled emotion image;

[0057] S51142206: Update the detected emotion using the updated facial image information.

[0058] By adopting the above technical solution, and by verifying the user's emotions, new detected emotions can be obtained based on the consistency between the eye orientation and the reference orientation, the viewed selected image, and viewing features. This reduces the influence of viewing features on the user's emotions during emotion analysis and improves the accuracy of game emotion detection.

[0059] Optionally, the method for verifying detected emotions further includes:

[0060] S51142211: Responding to the image detection information and preset call features to generate call image information;

[0061] S51142212: Responding to the call image information and preset mouth features to obtain the mouth shape;

[0062] S51142213: Identify the content of a call by the shape of the mouth;

[0063] S51142214: Responding to the mouth shape and the historical playtime to generate a call time point;

[0064] S51142215: In response to the call time point and the detected emotion, obtain the call emotion;

[0065] S51142216: Responding to the content of the call to identify the expected emotion;

[0066] S51142217: Remove the call emotion that is consistent with the estimated emotion from the detected emotion to form a new detected emotion.

[0067] Secondly, this application provides a multi-game integration platform, which adopts the following technical solution:

[0068] A multi-game integration platform, comprising:

[0069] The acquisition module is used to acquire historical games;

[0070] The memory is used to store a program for a game recommendation method based on big data analysis;

[0071] A processor is used to load, execute, and implement programs stored in memory.

[0072] In summary, this application includes at least one of the following beneficial technical effects:

[0073] 1. By analyzing the historical playtime of historical games, historical screen parameters and marked historical types are obtained. Then, by analyzing the historical screen parameters, marked historical types, and historical games, recommended games are obtained. This allows the system to automatically recommend games with similar gameplay to users, reducing the time users spend searching and improving the efficiency of users searching for games.

[0074] 2. By detecting users' emotional changes while playing games, we can obtain emotional images and extract marker colors from these images. This allows us to further understand users' interests in the game's color scheme or gameplay and extract corresponding color tones for game recommendations, thereby improving the accuracy of game recommendations.

[0075] 3. By verifying the user's emotions, new emotions can be detected based on the consistency between the eye direction and the reference direction, the image being viewed, viewing features, and the user's call status. This reduces the influence of viewing features on the user's emotions during emotion analysis and improves the accuracy of game emotion detection. Attached Figure Description

[0076] Figure 1 This is a flowchart of a game recommendation method based on big data analysis according to an embodiment of the present invention;

[0077] Figure 2 This is a flowchart of the method for obtaining the recommended game according to an embodiment of the present invention;

[0078] Figure 3 This is the method flow of the emotion detection verification method according to the embodiment of the present invention. Figure 1 ;

[0079] Figure 4 This is the method flow of the emotion detection verification method according to the embodiment of the present invention. Figure 2 . Detailed Implementation

[0080] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. All user data and other data involved in this application have been obtained with full consent, and the collection, use, and processing of related information comply with the relevant laws, regulations, and standards of the relevant countries and regions.

[0081] Reference Figure 1 This application discloses a game recommendation method based on big data analysis, including the following steps:

[0082] S1: Collect historical games.

[0083] The multi-game integration platform is a platform designed by technicians that integrates various games. Historical games refer to games that the user has previously played, which are retrieved from the multi-game integration platform system. In this embodiment, the multi-game integration platform includes a recommendation module, which the user can choose to enable. Before enabling the recommendation module, information such as the user's facial image and audio recordings will be displayed to facilitate game recommendations.

[0084] S2: Responds to historical games by retrieving historical playtime.

[0085] Historical playtime refers to the total time a user spends playing a historical game. It is calculated by retrieving the playtime of a historical game from a multi-game integration platform.

[0086] S3: Define the historical game with the longest playtime as the marked historical game.

[0087] A marked history game is a historical game with the longest playtime. The historical game with the longest playtime is defined as a marked history game.

[0088] S4: Responds to the marking of historical games by retrieving historical screen parameters and marking historical types.

[0089] Historical screen parameters refer to the screen parameters of a user playing a historical game. These are obtained by retrieving parameters such as contrast and brightness from the user's screen while playing the marked historical game from a multi-game integration platform. Marked historical type refers to the type of game being marked, retrieved from the marked historical game list. Marked historical types include card games, competitive games, and casual games, among others.

[0090] S5: Responds to the tagging of historical types, historical games, and historical screen parameters to obtain recommended games, and outputs the recommended games to the preset display area.

[0091] Recommended games refer to games recommended to users by the multi-game integration platform. The display area is the area on the multi-game integration platform designated by technical personnel for displaying recommended games. Recommended games are obtained by analyzing historical game types, historical games, and historical screen parameters, and then output to the display area.

[0092] Reference Figure 2 Methods to get recommended games include:

[0093] S51: Responds to history games and marking games to obtain a range of marking tones.

[0094] The marked color range refers to the range of colors displayed on the screen when a user plays the history game and the marked game. The marked color range is obtained by analyzing the history game and the marked game.

[0095] S52: Responds to tagging historical types to select similar games.

[0096] Similar games refer to games in a multi-game integration platform that are of the same type as those marked in the historical data. Games that are of the same type as those marked in the historical data are retrieved from the multi-game integration platform and identified as similar games.

[0097] S53: Responds to similar games by invoking a baseline display tone.

[0098] The baseline display tone refers to the tone of images displayed by similar games. The baseline display tone is obtained by retrieving images displayed by similar games from a multi-game integration platform and performing tone analysis. The tone analysis method is common knowledge to those skilled in the art and will not be elaborated here.

[0099] S54: Select recommended games by marking the tonal range and displaying the tonal range against a baseline.

[0100] Recommended games are those that are similar to the hues shown on a benchmark that includes the marked hue range.

[0101] Methods for obtaining the marked hue range include:

[0102] S5101: Captures the marking game screen of the marking game.

[0103] The marked game screen refers to the screen displayed by the game when the user plays the marked game. The marked game screen is obtained by collecting the screen displayed by the user when playing the marked game from a multi-game integration platform.

[0104] S5102: Responds to marking game screen to generate tone type.

[0105] Color tone type refers to the type of color tone displayed in the marked game screen. By analyzing the color tone of the marked game screen, the color tone type with the largest color tone range is taken as the color tone type. Color tone types include cool color tone and warm color tone.

[0106] S5103: Responds to color type, historical games, and marked historical games to detect games.

[0107] Detection games refer to historical games with the same color tone and type as those marked historical games. By selecting historical games with the same color tone and type from the historical games, the selected historical games and marked historical games are defined as detection games.

[0108] S5104: In response to detecting and marking game screens, generate a detection screen.

[0109] The detection screen refers to the screen displayed when a user plays the detection game and the screen displayed when the game is marked. The detection game screen is retrieved from the multi-game integration platform and used together with the marked game screen as the detection screen.

[0110] S5105: Responds to the detected screen to obtain the marked tonal range.

[0111] The marked color range is obtained by analyzing the detected image. In this embodiment, the marked color range is the color range that the user likes or dislikes.

[0112] Other methods for obtaining the marked hue range include:

[0113] S5111: Responds to game detection to obtain the game type.

[0114] Detecting game type refers to identifying the game's genre by retrieving the game genre from the game detection database. In this embodiment, a game may have one or more game genres. For example, a card-based competitive game.

[0115] S5112: Obtain the detection range by detecting the game type.

[0116] The marked detection range refers to the area used to select the game screen for tone detection. The marked detection range is retrieved from a preset detection lookup table based on the game type. The detection lookup table stores the marked detection ranges corresponding to different game types. The parameters in the detection lookup table are pre-set experimentally by those skilled in the art based on actual conditions and will not be elaborated here. In this embodiment, the type matched by the marked detection range is the first type of the game, for example: card battle games involve exchanging cards for combat, while competitive card games involve characters using card skills for combat.

[0117] For example: card games should be captured in a 65x65 pixel format with the center of the screen spread out, while competitive games should be captured in a 70x70 pixel format with the center of the screen spread out.

[0118] S5113: Responds to the detection screen and marks the detection range to obtain the detection selection screen.

[0119] The detection frame selection screen refers to the screen displayed after selecting the detection area by marking it. The screen obtained by cropping the detection area from the detection screen is used as the detection frame selection screen.

[0120] S5114: Responds to the detection of the selected area to generate a marker feature color.

[0121] The marker feature color refers to the color within the detection bounding box of the image. This marker feature color is obtained by identifying the detected bounding box. Image recognition technology is common knowledge to those skilled in the art and will not be elaborated upon here.

[0122] S5115: Responds to the marker feature color and a preset base color to obtain the marker hue parameter.

[0123] The reference color is a standard color set by technicians that has not been adjusted for parameters such as brightness and contrast. The marker hue parameter refers to the hue parameter presented by the marker feature color. The marker hue parameter is obtained by comparing and analyzing the marker feature color with the reference color. The method of hue parameter analysis is common knowledge to those skilled in the art and will not be elaborated here.

[0124] S5116: Responds to the marker hue parameter to generate a marker hue range.

[0125] The range of marked tones is obtained by combining the individual marked tone parameters.

[0126] Methods for obtaining the marker hue parameters include:

[0127] S5121: Extract a color from the marker colors that is inconsistent with the base color as a neutral color.

[0128] Neutral colors are those not included in the hue category, such as standard gray and brown. Neutral colors are selected from the base colors and those that do not conform to the base color.

[0129] S5122: Responds to historical screen parameters and neutral colors to obtain neutral tone parameters.

[0130] Neutral tone parameters refer to the hue of neutral colors displayed on the user's screen. Neutral tone parameters are obtained by analyzing historical screen parameters and neutral colors. The methods for analyzing the hue of neutral colors are common knowledge to those skilled in the art and will not be elaborated here.

[0131] S5123: Update the marker tone parameter using the neutral tone parameter.

[0132] A new marker tone parameter is obtained by adding the neutral tone parameter to the marker tone parameter.

[0133] Methods for generating marker feature colors also include:

[0134] S51141: Collect image detection information from the user.

[0135] Image detection information is obtained by capturing images of the user playing historical games using the on-screen camera.

[0136] S51142: Responds to image detection information to generate a mood-changing image.

[0137] Emotional change images refer to images in which users exhibit emotional changes. These images are obtained by performing facial expression recognition on the image detection information. The methods for facial expression recognition of emotions are common knowledge to those skilled in the art and will not be elaborated upon here.

[0138] S51143: The duration of emotional changes can be retrieved by comparing the emotional change scene with the historical playtime.

[0139] The duration of emotional change refers to the length of time a user experiences an emotional change while playing the game. This is determined by retrieving the duration corresponding to the emotional change screenshots from historical playtime data. In this embodiment, the duration of emotional change also includes the start and end times corresponding to the duration.

[0140] S51144: Collect and detect game interaction commands.

[0141] Interaction commands refer to the instructions output by the user to the game, which are retrieved from the multi-game integration platform. Interaction commands include instruction information output by devices such as keyboards, mice, and gamepads.

[0142] S51145: Retrieves interaction duration in response to interaction commands and historical play duration.

[0143] Interaction duration refers to the length of time it takes for a user to output an interaction command to the game. This is determined by retrieving the corresponding duration from historical playtime data. In this embodiment, the interaction duration includes the start and end times corresponding to the specified duration.

[0144] S51146: Respond to the interaction duration and the duration of emotion change to generate an emotion detection image, and generate a marker feature color by detecting the emotion image.

[0145] Emotion detection images refer to images that evoke emotions in users while playing games. These images are obtained by analyzing the duration of interaction and the duration of emotion changes, and then image analysis is performed on them to obtain marker feature colors.

[0146] Methods for generating images showing emotional changes include:

[0147] S511421: Responds to image detection information and preset facial features to obtain facial image information.

[0148] Facial features refer to the shape and color of a person's face as defined by technicians. Facial image information refers to the user's facial image, which is obtained by selecting the image corresponding to the facial features from the image detection information.

[0149] S511422: Update facial image information in response to interaction duration and emotion change duration.

[0150] The new facial image information is obtained by retrieving images within the overlapping timeframe of the interaction duration and the emotional change duration from various facial image information.

[0151] S511423: Response to facial image information to obtain detected emotion.

[0152] Emotion detection refers to the emotion displayed by facial image information. Emotions are detected by analyzing facial expressions. Methods for analyzing emotions through facial expressions are common knowledge to those skilled in the art and will not be elaborated upon here. For example, if a user laughs loudly and frequently outputs commands, the emotion displayed by the user is liking that color tone.

[0153] S511424: Select images showing emotional changes from the detected frame by detecting emotions.

[0154] The emotional change images are selected from the various detection frames at the corresponding time points.

[0155] Reference Figure 3 Verification methods for detecting emotions include:

[0156] S51142201: Responds to facial image information and preset eye features to obtain an eye image.

[0157] An eye image refers to an image of a user's eyes, which is obtained by retrieving the image corresponding to the eye features from facial image information.

[0158] S51142202: Obtain eye orientation from eye image.

[0159] Eye orientation refers to the direction in which a user's eyes are facing. Eye orientation is determined by analyzing eye images. Image recognition methods are common knowledge to those skilled in the art and will not be elaborated upon here.

[0160] S51142203: When the eye orientation is inconsistent with the preset reference orientation, the image of the viewing box is retrieved in response to the eye orientation and image detection information.

[0161] The baseline orientation is set by the technicians to indicate the direction in which the user's eyes view the game screen. When the eye orientation matches the baseline orientation, it indicates that the user is not looking at other objects, and the output continues to detect emotions.

[0162] The viewing frame image refers to the image of the object that the user's eyes are looking at. When the direction of the eyes is inconsistent with the reference direction, it means that the user is not looking at the game screen and is looking at other objects. In this case, the image of the object corresponding to the direction of the eyes is retrieved from the image detection information and used as the viewing frame image.

[0163] S51142204: When the selected image contains preset viewing features, respond to the viewing features, reference orientation, and eye orientation to obtain a labeled emotion image.

[0164] Viewing features are defined by technicians as characteristics of items that evoke emotional changes in users after viewing them. Viewing features include characteristics of items such as mobile phones and books. If the selected image does not contain viewing features, it indicates that the user has not viewed other items, and the system continues to output the detected emotion.

[0165] A labeled emotion image is an image in which a user experiences emotional changes while viewing other objects. When the image in the viewing frame contains viewing features, it indicates that the user is experiencing emotional changes while viewing other objects. In this case, the facial image information corresponding to the eye orientation that is inconsistent with the reference orientation is selected, and the selected facial image information containing viewing features is defined as a labeled emotion image.

[0166] S51142205: Responds to tagged emotion images to update facial image information.

[0167] New facial image information is obtained by removing labeled emotion images from facial image information.

[0168] S51142206: Update detected emotions using updated facial image information.

[0169] Re-execute S511423 to obtain new detected emotions.

[0170] Reference Figure 4 Verification methods for detecting emotions also include:

[0171] S51142211: Responds to image detection information and preset call features to generate call image information.

[0172] Call features are characteristics of user calls defined by technicians. Call image information refers to the image of a user during a call, which is selected from the image detection information to identify the image corresponding to the call features. This call image information includes images of the phone being held to the ear during a call, or images of the user using speakerphone. When the user is using speakerphone, the screen displayed on the phone is the call feature, such as the image indicating that the call has ended.

[0173] S51142212: Responds to call image information and preset mouth features to obtain mouth shape.

[0174] Mouth features are the shape and color of the mouth as defined by technicians. Mouth shape refers to the shape of the user's mouth during a call, which is determined by selecting the shape of the image corresponding to the mouth features from the call image information.

[0175] S51142213: Identify call content by mouth shape.

[0176] The content of the call refers to the estimated content of the user's conversation, which is obtained through lip reading based on the shape of the mouth. The method of lip reading based on images is common knowledge to those skilled in the art and will not be elaborated here.

[0177] S51142214: Responds to mouth shape and historical playtime to generate call timestamps.

[0178] The call time point refers to the time when the user makes a call. It is determined by retrieving the time point when the user's mouth shape changes during the call from the historical play duration.

[0179] S51142215: Responds to the call time point and detects emotions to obtain the call emotion.

[0180] Call emotion refers to the detected emotion at a specific point in time during a call. The call emotion is determined by retrieving the emotion corresponding to the specific point in time from various detected emotions.

[0181] S51142216: Respond to the content of the call to identify the expected emotion.

[0182] Estimated emotion refers to the emotion a user anticipates during a call, which is obtained by identifying the content of the call. Methods for emotion recognition through text are common knowledge in the field and will not be elaborated upon here.

[0183] S51142217: Remove the call emotions that are consistent with the predicted emotions from the detected emotions to form new detected emotions.

[0184] New detected emotions are obtained by removing those from the detected emotions that are consistent with the predicted emotions.

[0185] Further analytical methods for detecting sentiment and recommending games include:

[0186] S5114201: Respond to recommended games to obtain labeled recommended sentiment.

[0187] The labeled recommendation emotion refers to the emotion of users when playing recommended games. Refer to S511423 to S51142217 to obtain the labeled recommendation emotion through the game screen of recommended games.

[0188] S5114202: Collect game reviews and images of recommended games.

[0189] Game reviews refer to comments on recommended games, while recommended game images refer to gameplay footage of users playing recommended games. Game reviews and recommended game images are retrieved from a multi-game integration platform.

[0190] S5114203: Responds to recommended game images and game reviews to obtain actual gameplay reviews.

[0191] Actual gameplay reviews refer to game reviews that correspond to the game footage played by the user. By analyzing recommended game images to obtain text content, and then selecting game reviews based on the overlap of keywords between the text content and game reviews, these reviews are considered actual gameplay reviews.

[0192] S5114204: Respond to actual gameplay reviews to generate estimated gameplay sentiment.

[0193] Estimated player sentiment refers to the player sentiment reflected in actual gameplay reviews. This is achieved by analyzing the text of actual gameplay reviews. The methods for analyzing sentiment through text are common knowledge in the field and will not be elaborated upon here.

[0194] S5114205: Update game reviews in response to anticipated player sentiment and flagged recommended sentiment.

[0195] By falsely labeling reviews that do not match the predicted play mood, new game reviews can be generated, making it easier for other players to analyze the game content through these reviews.

[0196] Based on the same inventive concept, embodiments of the present invention provide a multi-game integration platform, including:

[0197] The acquisition module is used to acquire historical games, marked game screens, image detection information, interaction commands, game reviews, and recommended game images;

[0198] The memory is used to store a program for a game recommendation method based on big data analysis;

[0199] A processor is used to load, execute, and implement programs stored in memory.

[0200] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0201] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A game recommendation method based on big data analysis, characterized in that, include: S1: Collect historical games; S2: Respond to the historical game to retrieve the historical playtime; S3: Define the historical game with the longest historical playtime as the marked historical game; S4: Retrieve the history marking game to retrieve history screen parameters and history marking type; S5: In response to the marked history type, the history game, and the history screen parameters, a recommended game is obtained, and the recommended game is output to a preset display area; The methods for obtaining the recommended games include: S51: Responding to the historical game and the marked historical game to obtain a marked hue range; S52: In response to the stated tag history type, select similar games; S53: Respond to the similar game by invoking a baseline display hue; S54: Select the recommended game by comparing the marked hue range with the reference display hue; The methods for obtaining the marked hue range include: S5101: Collect the marked game screen of the marked history game; S5102: Responding to the marked game screen to generate a color tone type; S5103: Detect game in response to the tone type, the history game, and the marked history game; S5104: In response to the detected game and the marked game screen, generate a detection screen; S5105: Respond to the detection screen to obtain the marked tone range; Also includes: S5111: In response to the detected game, obtain the detected game type; S5112: Obtain the marked detection range by detecting the game type; S5113: Responding to the detection screen and the marked detection range to obtain a detection box selection screen; S5114: Respond to the detection frame selection screen to generate a marker feature color; S5115: Responding to the marker feature color and the preset reference color to obtain the marker hue parameter; S5116: Responding to the marker hue parameter to generate the marker hue range; The method for generating the marker feature color further includes: S51141: Collect image detection information from the user; S51142: Responding to the image detection information to generate an emotional change image; S51143: The duration of emotional changes is retrieved by comparing the emotional change images with the historical playtime. S51144: Collect the interaction commands of the detected game; S51145: Retrieve the interaction duration in response to the interaction command and the historical play duration; S51146: Responding to the interaction duration and the emotion change duration to generate an emotion detection screen, and using the emotion detection screen to generate the marker feature color.

2. The game recommendation method based on big data analysis according to claim 1, characterized in that, The methods for obtaining the marker hue parameters include: S5121: Extract a color from the marked feature colors that is inconsistent with the reference color as a neutral color; S5122: Responding to the historical screen parameters and the neutral color to obtain neutral tone parameters; S5123: Update the marker tone parameter using the neutral tone parameter.

3. The game recommendation method based on big data analysis according to claim 1, characterized in that, The methods for generating the aforementioned emotional change images include: S511421: Responding to the image detection information and preset facial features to obtain facial image information; S511422: Update the facial image information in response to the interaction duration and the emotion change duration; S511423: Response to the facial image information to obtain detected emotion; S511424: Select the emotion change image from the detection frame screen by detecting the emotion.

4. The game recommendation method based on big data analysis according to claim 3, characterized in that, The method for verifying detected emotions includes: S51142201: Obtain an eye image in response to the facial image information and preset eye features; S51142202: Obtain the eye orientation from the eye image; S51142203: When the eye orientation is inconsistent with the preset reference orientation, the image to be viewed is retrieved in response to the eye orientation and the image detection information; S51142204: When the selected image contains a preset viewing feature, a labeled emotion image is obtained in response to the viewing feature, the reference orientation, and the eye orientation; S51142205: Update the facial image information in response to the labeled emotion image; S51142206: Update the detected emotion using the updated facial image information.

5. The game recommendation method based on big data analysis according to claim 4, characterized in that, The method for verifying detected emotions also includes: S51142211: Responding to the image detection information and preset call features to generate call image information; S51142212: Responding to the call image information and preset mouth features to obtain the mouth shape; S51142213: Identify the content of a call by the shape of the mouth; S51142214: Responding to the mouth shape and the historical playtime to generate a call time point; S51142215: In response to the call time point and the detected emotion, obtain the call emotion; S51142216: Responding to the content of the call to identify the expected emotion; S51142217: Remove the call emotion that is consistent with the estimated emotion from the detected emotion to form a new detected emotion.

6. A multi-game integration platform, characterized in that, include: The acquisition module is used to acquire historical games; A memory for storing a program of a game recommendation method based on big data analysis as described in any one of claims 1 to 5; A processor is used to load, execute, and implement programs stored in memory.

Images