Novel blind box lottery drawing probability dynamic adjustment algorithm

Abstract

The invention belongs to the technical field of probability adjustment, and discloses a novel blind box lottery drawing probability dynamic adjustment algorithm. Comprising the following steps: S1, grading articles in a blind box, and setting an initial probability value for each grade of articles; s2, acquiring the accumulated purchase times of the user and the purchase times in unit time, and dynamically adjusting the initial probability value of each grade of article in the blind box according to the accumulated purchase times of the user and the purchase times in unit time, so as to obtain the adjustment probability value of each grade of article; according to the method, through dynamic adjustment of the initial probability value and a bottom guarantee mechanism, the winning probability can be gradually increased when the user continuously does not hold the rare articles, disappropriation and loss caused by the fact that the user does not hold the rare articles for a long time are prevented, the user tag is generated by obtaining the historical purchase record of the user, the probability value of each grade of articles is adjusted according to the tag, and the user experience is improved. And accurate matching is realized.

Description

Technical Field

[0001] The present invention relates to the field of probability adjustment technology, and more specifically, to a novel blind box prize drawing probability dynamic adjustment algorithm. Background Art

[0002] As a new user purchasing method, blind box gameplay has been widely used in e-commerce, games and virtual goods transactions in recent years. However, the traditional blind box system only determines the winning probability of an item through a simple random number generation mechanism, and cannot dynamically adjust the winning probability based on user behavior or user tags and historical purchase times, and the probability control of rare items is not precise enough. Therefore, the current blind box system has obvious deficiencies in user experience, winning probability control and item distribution balance.

[0003] In view of this, the present invention proposes a novel blind box prize drawing probability dynamic adjustment algorithm to solve the above problems. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a new blind box prize probability dynamic adjustment algorithm, comprising the following steps:

[0005] S1. Classify the items in the blind box into different levels and set the initial probability value for each level of items;

[0006] S2. Obtain the user's cumulative purchase times and the number of purchases per unit time, and dynamically adjust the initial probability value of each level of items in the blind box according to the user's cumulative purchase times and the number of purchases per unit time, and then obtain the adjusted probability value of each level of items;

[0007] S3. Obtain the user tag of the user, and modify the adjusted probability value of the item corresponding to each level according to the user tag, so as to obtain the final probability value of the item corresponding to each level of the user.

[0008] Furthermore, the method of classifying the items in the blind box and setting the initial probability value for each level of items includes:

[0009] The items in the blind box are classified into the following levels in ascending order of their cost: common items, rare items, and super rare items;

[0010] Set the initial probability value of items corresponding to each level, and the initial probability values ​​of common items, rare items, and super rare items decrease in turn.

[0011] Furthermore, the method of obtaining the user's cumulative purchase times and the purchase times per unit time, and dynamically adjusting the initial probability value of each level of items in the blind box according to the user's cumulative purchase times and the purchase times per unit time, and then obtaining the adjusted probability value of each level of items, includes:

[0012] Establish a cumulative purchase number-probability adjustment model and a purchase number-probability adjustment model per unit time;

[0013] Obtain the cumulative number of purchases and the number of purchases per unit time from the user's account;

[0014] The user's cumulative purchase times are input into the cumulative purchase times-probability adjustment model, and the user's purchase times per unit time are input into the purchase times per unit time-probability adjustment model;

[0015] The cumulative purchase count-probability adjustment model and the probability value output by the probability adjustment model are linearly superimposed to obtain the adjusted probability value of each level of items.

[0016] Furthermore, the method of establishing the cumulative purchase frequency-probability adjustment model includes:

[0017] W is preset as the probability value output by the cumulative number of purchases-probability adjustment model, N is preset as the initial probability value, α is preset as the slope of the cumulative number of purchases-probability adjustment model, β is the first bias term (indicating the influence of the cumulative number of purchases), and the expression of the cumulative number of purchases-probability adjustment model is: W=α·N+β.

[0018] Furthermore, the method of establishing the purchase frequency-probability adjustment model per unit time includes:

[0019] Preset W` as the probability value output by the purchase number-probability adjustment model per unit time, preset N` as the initial probability value, preset α` as the slope of the purchase number-probability adjustment model per unit time, β` as the second bias term (indicating the influence of the purchase number per unit time), and then the expression of the purchase number-probability adjustment model per unit time is: W`=α`·N`+β`.

[0020] Furthermore, the method of linearly superimposing the cumulative number of purchases-probability adjustment model and the probability value output by the probability adjustment model includes:

[0021] Preset Wt as the adjustment probability value, preset τ1 as the first weight, preset τ2 as the second weight, and then the expression for linear superposition is: Wt=τ1·w+τ2·w`, where τ1+τ2=1.

[0022] Furthermore, in S2, when the user reaches a preset cumulative purchase count threshold within a preset time threshold and does not obtain a rare item, a guaranteed mechanism is triggered, and the guaranteed mechanism is that in the next lottery, the probability of the user winning a rare item is 100%.

[0023] Furthermore, after the minimum guarantee mechanism is triggered, the user's adjusted probability value is reset, and then the user's adjusted probability value is reset to an initial probability value.

[0024] Furthermore, the method of obtaining the user tag of the user and modifying the adjusted probability value of the item corresponding to each level according to the user tag of the user, thereby obtaining the final probability value of the item corresponding to each level of the user includes:

[0025] Obtain the user's historical purchase records, obtain the type of items that the user purchased most based on the user's historical purchase records, and use this item type as the user's user tag;

[0026] Classify the items in the blind box until each item in the blind box has a corresponding item type;

[0027] Establish a mapping relationship between item types and user tags;

[0028] The adjusted probability value of the item type corresponding to each user level is modified according to the mapping relationship, and then the final probability value of the item corresponding to each user level is obtained.

[0029] Furthermore, the method of modifying the adjusted probability value of the item type corresponding to each user level according to the mapping relationship includes:

[0030] Get the items and item types corresponding to each level;

[0031] For the adjusted probability values ​​of items with mapping relationships within each level of the user, the preset weighted ratio Q is used to modify the adjusted probability values, thereby obtaining the final probability values ​​of the items corresponding to each level of the user.

[0032] The technical effects and advantages of a novel blind box prize-winning probability dynamic adjustment algorithm of the present invention are as follows:

[0033] 1. Through the dynamic adjustment of the initial probability value and the guarantee mechanism, the winning probability of users can be gradually increased when they have not won rare items for consecutive times, thus preventing users from being disappointed and leaving due to long-term failure to win rare items, enhancing the user's sense of participation, and improving the user experience and satisfaction; the probability distribution of items can be dynamically adjusted according to user tags, so that users have the motivation to continue to participate; the probability of each item can be accurately controlled to ensure that rare items appear within a reasonable range; and the winning probability distribution of suitable items can also be provided to users according to their user tags, thereby bringing a more personalized experience that meets user needs and enhances user loyalty. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a flow chart of a novel blind box prize-winning probability dynamic adjustment algorithm of the present invention;

[0035] Figure 2 It is a structural schematic diagram of the present invention;

[0036] Figure 3 It is a schematic diagram of the blind box prize drawing process of the present invention. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] Example 1

[0039] See also Figure 1 As shown, a novel blind box prize-winning probability dynamic adjustment algorithm described in this embodiment includes the following steps:

[0040] S1. Classify the items in the blind box into different levels and set the initial probability value for each level of items;

[0041] S2. Obtain the user's cumulative purchase times and the number of purchases per unit time, and dynamically adjust the initial probability value of each level of items in the blind box according to the user's cumulative purchase times and the number of purchases per unit time, and then obtain the adjusted probability value of each level of items;

[0042] S3. Obtain the user tag of the user, and modify the adjusted probability value of the item corresponding to each level according to the user tag, so as to obtain the final probability value of the item corresponding to each level of the user.

[0043] Furthermore, the items in the blind box are graded and the initial probability value is set for each grade of items, including:

[0044] The items in the blind box are classified into the following levels in ascending order of their cost: common items, rare items, and super rare items;

[0045] Set the initial probability value of items corresponding to each level, and the initial probability values ​​of common items, rare items, and super rare items decrease in sequence, where the sum of the initial probability values ​​of items of all levels is 1.

[0046] Specifically, the items in the blind box can be divided into different levels such as common, rare and super rare. Set the initial probability value for each level of items: common items: 70%; rare items: 25%; super rare items: 5%; you can also use a pseudo-random number generator to generate a random number for the prize, and determine the winning level based on the probability threshold of the item level; the pseudo-random number generation algorithm uses Mersenne-Twi ster (Mersenne rotation method) or linear congruential method (LCG), and then, you can ensure the randomness of the generated results.

[0047] It should be noted that by dividing the items in the blind box into levels and setting initial probability values ​​for items in each level, the diversity and appeal of the blind box are ensured; users can choose different blind boxes according to their preferences and budgets, which increases the fun and anticipation of purchase; by adopting a pseudo-random number generator (such as the Mersenne-Twi ster or linear congruential method) to generate random numbers for the draw, the randomness and unpredictability of the draw results are ensured; this helps to maintain the mystery and appeal of the blind box and prevent cheating and fraud.

[0048] Furthermore, the cumulative number of purchases and the number of purchases per unit time of the user are obtained, and the initial probability value of each level of items in the blind box is dynamically adjusted according to the cumulative number of purchases and the number of purchases per unit time of the user, and then the adjusted probability value of each level of items is obtained, including:

[0049] Establish a cumulative purchase number-probability adjustment model and a purchase number-probability adjustment model per unit time;

[0050] Obtain the cumulative number of purchases and the number of purchases per unit time from the user's account (the length of the unit time is set by relevant personnel based on actual conditions);

[0051] The user's cumulative purchase times are input into the cumulative purchase times-probability adjustment model, and the user's purchase times per unit time are input into the purchase times per unit time-probability adjustment model;

[0052] The cumulative purchase count-probability adjustment model and the probability value output by the probability adjustment model are linearly superimposed to obtain the adjusted probability value of each level of items.

[0053] Furthermore, the method of establishing the cumulative purchase frequency-probability adjustment model includes:

[0054] W is preset as the probability value output by the cumulative number of purchases-probability adjustment model, N is preset as the initial probability value, α is preset as the slope of the cumulative number of purchases-probability adjustment model, β is the first bias term (indicating the influence of the cumulative number of purchases), and the expression of the cumulative number of purchases-probability adjustment model is: W=α·N+β.

[0055] Furthermore, the method of establishing a purchase frequency-probability adjustment model per unit time includes:

[0056] Preset W` as the probability value output by the purchase number-probability adjustment model per unit time, preset N` as the initial probability value, preset α` as the slope of the purchase number-probability adjustment model per unit time, β` as the second bias term (indicating the influence of the purchase number per unit time), and then the expression of the purchase number-probability adjustment model per unit time is: W`=α`·N`+β`.

[0057] Furthermore, the method of linearly superimposing the cumulative number of purchases-probability adjustment model and the probability value output by the probability adjustment model includes:

[0058] Preset Wt as the adjustment probability value, preset τ1 as the first weight, preset τ2 as the second weight, and then the expression for linear superposition is: Wt=τ1·w+τ2·w`, where τ1+τ2=1.

[0059] Specifically, the more times a user purchases, the higher the probability of obtaining rare items. By adjusting the initial probability value, the probability of obtaining rare items will gradually increase for users who consume blind boxes frequently and accumulate a large number of blind boxes.

[0060] It should be noted that personalized recommendations are achieved by dynamically adjusting the probability values ​​of items of each level based on the user's cumulative purchase times and the number of purchases per unit time; this can more accurately reflect the user's needs and preferences, and improve the satisfaction and fairness of winning.

[0061] Furthermore, in S2, when the user reaches a preset cumulative purchase count threshold within a preset time threshold and does not obtain a rare item, a guaranteed mechanism is triggered. The guaranteed mechanism is that in the next lottery, the user has a 100% chance of winning a rare item. The time threshold is set by relevant personnel based on actual conditions, such as 15 days.

[0062] Furthermore, after the minimum guarantee mechanism is triggered, the user's adjusted probability value is reset, and then the user's adjusted probability value is reset to the initial probability value.

[0063] It should be noted that the minimum guarantee mechanism can prevent users from not obtaining rare items for a long time; by introducing the minimum guarantee mechanism, it ensures that when users reach a certain number of purchases and do not obtain rare items, they will win rare items in the next draw; it can effectively avoid user frustration and loss, and enhance user loyalty and stickiness; after the minimum guarantee mechanism is triggered, the user's adjusted probability value is reset to the initial probability value, avoiding probability imbalance caused by continuous winnings; it can ensure the stability and sustainability of the algorithm, and provide strong support for long-term operations.

[0064] Furthermore, a method of obtaining a user tag of a user, and modifying the adjusted probability value of an item corresponding to each level according to the user tag of the user, and then obtaining a final probability value of an item corresponding to each level of the user includes:

[0065] Obtain the user's historical purchase records, obtain the type of items that the user purchased most based on the user's historical purchase records, and use this item type as the user's user tag;

[0066] Classify the items in the blind box until each item in the blind box has a corresponding item type;

[0067] Establish a mapping relationship between item types and user tags;

[0068] The adjusted probability value of the item type corresponding to each user level is modified according to the mapping relationship, and then the final probability value of the item corresponding to each user level is obtained.

[0069] It should be noted that accurate matching is achieved by obtaining the user's historical purchase records to generate user tags and adjusting the probability values ​​of items at each level according to the tags; it can more accurately meet the needs of users and improve the targetedness and satisfaction of winning the prizes by adjusting the probability values ​​of items with mapping relationships within each level.

[0070] Furthermore, the method of modifying the adjusted probability value of the item type corresponding to each level of the user according to the mapping relationship includes:

[0071] Get the items and item types corresponding to each level;

[0072] For the adjusted probability values ​​of items with mapping relationships within each level of the user, the preset weighted ratio Q is used to modify the adjusted probability values, thereby obtaining the final probability values ​​of the items corresponding to each level of the user.

[0073] Specifically, for example, there are prizes A, B, and C in ordinary items; when creating prizes, it is necessary to map user tags to prizes A, B, and C respectively; for example, the tag of prize A is: chasing stars; the tag of prize B is: two-dimensional; the tag of prize C is: furniture; then, when user A draws the prize, according to user A's daily behavior records on the platform (user tags), it is obtained that user A has a preference for "chasing stars", so when user A draws the prize, the weight of prize A in ordinary items is dynamically adjusted; for example, the winning weight of prize A is adjusted from 1 to 5, so that when users hit ordinary items, they are more likely to win prize A;

[0074] It should be noted that the process of winning the corresponding blind box when the user purchases the blind box can be referred to Figure 3 ;

[0075] In this embodiment, through the dynamic adjustment of the initial probability value and the guarantee mechanism, the winning probability of the user can be gradually improved when the user fails to win the rare item continuously, thereby preventing the disappointment and loss of users caused by not winning the rare item for a long time, enhancing the user's sense of participation, and improving the user experience and satisfaction; the probability distribution of items can be dynamically adjusted according to user tags, so that users have the motivation to continue to participate. For example, the winning probability of high-frequency or high-consumption users will increase with changes in behavior, thereby increasing their willingness to continue purchasing and improving user stickiness; the probability of each item can be accurately controlled to ensure that rare items appear within a reasonable range; the winning probability distribution of suitable items can be provided to users according to their user tags, thereby bringing a personalized experience that is more in line with user needs and enhancing user loyalty.

[0076] Example 2

[0077] See also Figure 2 As shown, a novel blind box prize-winning probability dynamic adjustment system described in this embodiment includes:

[0078] The basic probability allocation module is used to classify the items in the blind box and set the initial probability value for each level of items;

[0079] An adaptive adjustment module is used to obtain the user's cumulative purchase times and the number of purchases per unit time, and dynamically adjust the initial probability value of each level of items in the blind box according to the user's cumulative purchase times and the number of purchases per unit time, thereby obtaining the adjusted probability value of each level of items;

[0080] The user behavior analysis module is used to obtain the user's user tag and modify the adjusted probability value of the item corresponding to each level according to the user's user tag, thereby obtaining the final probability value of the item corresponding to each level of the user.

[0081] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0082] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0083] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

[0084] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A new blind box prize probability dynamic adjustment algorithm, characterized in that: The following steps are involved: S1. Classify the items in the blind box into different levels and set the initial probability value for each level of items; S2. Obtain the user's cumulative purchase times and the number of purchases per unit time, and dynamically adjust the initial probability value of each level of items in the blind box according to the user's cumulative purchase times and the number of purchases per unit time, and then obtain the adjusted probability value of each level of items; S3. Obtain the user tag of the user, and modify the adjusted probability value of the item corresponding to each level according to the user tag, so as to obtain the final probability value of the item corresponding to each level of the user.

2. According to a new type of blind box prize probability dynamic adjustment algorithm according to claim 1, it is characterized in that: The method of classifying the items in the blind box and setting the initial probability value for each level of items includes: The items in the blind box are classified into the following levels in ascending order of their cost: common items, rare items, and super rare items; Set the initial probability value of items corresponding to each level, and the initial probability values ​​of common items, rare items, and super rare items decrease in turn.

3. According to claim 1, a new type of blind box prize probability dynamic adjustment algorithm is characterized in that: The method of obtaining the user's cumulative purchase times and the purchase times per unit time, and dynamically adjusting the initial probability value of each level of items in the blind box according to the user's cumulative purchase times and the purchase times per unit time, and then obtaining the adjusted probability value of each level of items, includes: Establish a cumulative purchase count-probability adjustment model and a purchase count-probability adjustment model per unit time; Obtain the cumulative number of purchases and the number of purchases per unit time from the user's account; The user's cumulative purchase times are input into the cumulative purchase times-probability adjustment model, and the user's purchase times per unit time are input into the purchase times per unit time-probability adjustment model; The cumulative purchase count-probability adjustment model and the probability value output by the probability adjustment model are linearly superimposed to obtain the adjusted probability value of each level of items.

4. A novel blind box prize-winning probability dynamic adjustment algorithm according to claim 3, characterized in that: The method of establishing the cumulative purchase frequency-probability adjustment model includes: W is preset as the probability value output by the cumulative number of purchases-probability adjustment model, N is preset as the initial probability value, α is preset as the slope of the cumulative number of purchases-probability adjustment model, β is the first bias term, and the expression of the cumulative number of purchases-probability adjustment model is: W=α·N+β.

5. According to claim 3, a new type of blind box prize probability dynamic adjustment algorithm is characterized in that: The method of establishing the purchase frequency-probability adjustment model per unit time includes: Preset W` as the probability value output by the purchase number-probability adjustment model per unit time, preset N` as the initial probability value, preset α` as the slope of the purchase number-probability adjustment model per unit time, β` as the second bias term, and then the expression of the purchase number-probability adjustment model per unit time is: W`=α`·N`+β`.

6. A novel blind box prize-winning probability dynamic adjustment algorithm according to claim 3, characterized in that: The method of linearly superimposing the cumulative purchase times-probability adjustment model and the probability value output by the probability adjustment model includes: Wt is preset as the adjustment probability value, τ1 is preset as the first weight, τ2 is preset as the second weight, and the expression for linear superposition is: Wt=τ1·w+τ2·w`.

7. A novel blind box prize-winning probability dynamic adjustment algorithm according to claim 2, characterized in that: In S2, when the user reaches a preset cumulative purchase count threshold within a preset time threshold and does not obtain a rare item, a guaranteed mechanism is triggered, and the guaranteed mechanism is that in the next lottery, the probability of the user winning a rare item is 100%.

8. According to claim 7, a new type of blind box prize probability dynamic adjustment algorithm is characterized in that: After the minimum guarantee mechanism is triggered, the user's adjusted probability value is reset, and then the user's adjusted probability value is reset to the initial probability value.

9. A novel blind box prize-winning probability dynamic adjustment algorithm according to claim 1, characterized in that: The method of obtaining the user tag of the user and modifying the adjusted probability value of the item corresponding to each level according to the user tag of the user, thereby obtaining the final probability value of the item corresponding to each level of the user, includes: Obtain the user's historical purchase records, obtain the type of items that the user purchased most based on the user's historical purchase records, and use this item type as the user's user tag; Classify the items in the blind box until each item in the blind box has a corresponding item type; Establish a mapping relationship between item types and user tags; The adjusted probability value of the item type corresponding to each user level is modified according to the mapping relationship, and then the final probability value of the item corresponding to each user level is obtained.

10. A novel blind box prize-winning probability dynamic adjustment algorithm according to claim 9, characterized in that: The method of modifying the adjusted probability value of the item type corresponding to each level of the user according to the mapping relationship includes: Get the items and item types corresponding to each level; For the adjusted probability values ​​of items with mapping relationships within each level of the user, the preset weighted ratio Q is used to modify the adjusted probability values, thereby obtaining the final probability values ​​of the items corresponding to each level of the user.

Images