Virtual item operation method and device based on emotion label and electronic device

Abstract

Embodiments of the present disclosure disclose a virtual item operation method and device based on emotion labels and an electronic device. A specific implementation of the method includes: performing data preprocessing on a real-time service data sequence; determining an observation window according to the preprocessed service data sequence; determining an emotion label according to a target service data sequence; in response to determining that the emotion label is a negative emotion label, generating negative adjustment control strategy information for a virtual item according to current control strategy information corresponding to the virtual item and the emotion label; in response to determining that the emotion label is a positive emotion label, generating positive adjustment control strategy information for the virtual item according to the current control strategy information and the emotion label; and performing virtual item regulation on the virtual item according to the current control strategy information or updated control strategy information. The implementation effectively prevents and protects abnormal operations, avoids possible data risks, and ensures device security.

Description

Technical Field

[0001] The embodiments disclosed herein relate to the field of computer technology, and more specifically to a method, apparatus, and electronic device for operating virtual items based on emotion tags. Background Technology

[0002] Operations on virtual items (such as acquiring and updating them) often require frequent execution, resulting in a massive number of related operations per unit of time. Manual anomaly control methods for these frequent virtual item operations struggle to comprehensively and effectively detect abnormal operations, hindering precise and effective control. This is especially true when abnormal operations are detected, as the inability to effectively prevent and protect against them can lead to data and device security risks.

[0003] The information disclosed in this background section is only intended to enhance the understanding of the background of the inventive concept, and therefore may contain information that does not form prior art known to those skilled in the art. Summary of the Invention

[0004] The summary portion of this disclosure is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description portion. This summary portion is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0005] Some embodiments of this disclosure propose methods, apparatuses, and electronic devices for operating virtual items based on emotion tags to address one or more of the technical problems mentioned in the background section above.

[0006] In a first aspect, some embodiments of this disclosure provide a virtual item operation method based on sentiment tags. The method includes: collecting a real-time business data sequence for the virtual item; performing data preprocessing on the real-time business data sequence to obtain a preprocessed business data sequence; determining an observation window based on the preprocessed business data sequence; determining a sentiment tag based on a target business data sequence, wherein the target business data sequence is the preprocessed business data within the observation window in the preprocessed business data sequence, and the sentiment tag characterizes the risk control tendency for the virtual item; in response to determining that the sentiment tag is a negative sentiment tag, generating negative adjustment control strategy information for the virtual item based on the current control strategy information corresponding to the virtual item and the sentiment tag, as updated control strategy information; in response to... Upon determining that the aforementioned emotion tag is a positive emotion tag, positive adjustment control strategy information for the aforementioned virtual item is generated based on the aforementioned current control strategy information and the aforementioned emotion tag, serving as the updated control strategy information. Virtual item regulation is then performed on the aforementioned virtual item based on the aforementioned current control strategy information or the aforementioned updated control strategy information, including: determining abnormal control information based on the aforementioned current control strategy information or the aforementioned updated control strategy information, as well as the emotion tag and the operation anomaly prediction model; in response to determining that the abnormal control information indicates a control anomaly in the current control strategy information or the updated control strategy information, determining the control instruction corresponding to the abnormal control information in the abnormal control decision tree; and locking the authorized device related to the virtual item, or activating a firewall to block operations related to the virtual item, based on the control instruction.

[0007] Secondly, some embodiments of this disclosure provide a virtual item operation device based on emotion tags. The device includes: a data acquisition unit configured to acquire real-time business data sequences for virtual items; a data preprocessing unit configured to preprocess the real-time business data sequences to obtain preprocessed business data sequences; a first determination unit configured to determine an observation window based on the preprocessed business data sequences; a second determination unit configured to determine an emotion tag based on a target business data sequence, wherein the target business data sequence is the preprocessed business data within the observation window in the preprocessed business data sequence, and the emotion tag represents the risk control tendency for the virtual item; and a first generation unit configured to, in response to determining that the emotion tag is a negative emotion tag, generate negative adjustment control strategy information for the virtual item based on the current control strategy information corresponding to the virtual item and the emotion tag, and perform... The system includes: a second generation unit configured to, in response to determining that the aforementioned emotion tag is a positive emotion tag, generate positive adjustment control strategy information for the aforementioned virtual item based on the aforementioned current control strategy information and the aforementioned emotion tag, as the updated control strategy information; and a virtual item control unit configured to control the aforementioned virtual item based on the aforementioned current control strategy information or the aforementioned updated control strategy information, including: determining abnormal control information based on the aforementioned current control strategy information or the aforementioned updated control strategy information, as well as the emotion tag and the operation anomaly prediction model; in response to determining that the abnormal control information indicates that the current control strategy information or the updated control strategy information has a control anomaly, determining the control instruction corresponding to the abnormal control information in the abnormal control decision tree; and locking the authorized device related to the virtual item or activating the firewall to block the virtual item-related operations based on the control instruction.

[0008] Thirdly, some embodiments of this disclosure provide an electronic device, including: one or more processors; and a storage device having one or more programs stored thereon, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation of the first aspect above.

[0009] Fourthly, some embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method described in any of the implementations of the first aspect above.

[0010] The above embodiments of this disclosure have the following beneficial effects: The virtual item operation method based on emotion tags in some embodiments of this disclosure effectively prevents and protects against abnormal operations, avoiding potential data risks and ensuring device security. Specifically, the reason for the above problems is that, for frequent virtual item operations, using methods such as manual anomaly control makes it difficult to comprehensively and effectively detect abnormal virtual item operations, thus making it impossible to accurately and effectively control and schedule virtual items. Especially when abnormal virtual item operations exist, it is impossible to effectively prevent and protect against abnormal operations, which may lead to data risks and device security risks. Based on this, the virtual item operation method based on emotion tags in some embodiments of this disclosure first collects real-time business data sequences for virtual items to obtain real-time data for virtual items. Second, the real-time business data sequences are preprocessed to obtain preprocessed business data sequences. Data preprocessing improves data quality and eliminates noise that low-quality data may cause to subsequent results. Then, an observation window is determined based on the preprocessed business data sequences. In practice, an observation window is dynamically determined by combining the preprocessed business data sequence to facilitate the subsequent capture of abnormal control tendencies. Further, a sentiment label is determined based on the target business data sequence, where the target business data sequence is the preprocessed business data within the observation window from the aforementioned preprocessed business data sequence, and the sentiment label represents the risk control tendency towards the aforementioned virtual item. This, combined with the preprocessed business data within the observation window, quantifies the risk control tendency towards the virtual item through the sentiment label. Furthermore, in response to determining that the sentiment label is a negative sentiment label, negative adjustment control strategy information for the aforementioned virtual item is generated based on the current control strategy information corresponding to the virtual item and the sentiment label, serving as the updated control strategy information. Subsequently, in response to determining that the sentiment label is a positive sentiment label, positive adjustment control strategy information for the aforementioned virtual item is generated based on the current control strategy information and the sentiment label, serving as the updated control strategy information. Based on this, the current control strategy information and emotion tags are combined to update the corresponding control strategy information. Finally, according to the current control strategy information or the updated control strategy information, the virtual item is regulated, including: determining abnormal control information based on the current control strategy information or the updated control strategy information, as well as the emotion tags and the operation anomaly prediction model; in response to determining that the abnormal control information indicates that there is a control anomaly in the current control strategy information or the updated control strategy information, determining the control instruction corresponding to the abnormal control information in the abnormal control decision tree; and locking the authorized device related to the virtual item or activating the firewall to block the operation related to the virtual item according to the control instruction.In particular, by locking the trusted device or activating the firewall to block abnormal virtual item operations, abnormal operations can be effectively prevented and protected, avoiding potential data risks and ensuring device security. Attached Figure Description

[0011] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and elements are not necessarily drawn to scale.

[0012] Figure 1 This is a flowchart of some embodiments of the virtual item operation method based on emotion tags according to this disclosure;

[0013] Figure 2 This is a schematic diagram illustrating the process of determining the target business data sequence;

[0014] Figure 3 This is a schematic diagram of the structure of some embodiments of the virtual item manipulation device based on emotion tags according to the present disclosure;

[0015] Figure 4 This is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. Detailed Implementation

[0016] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0017] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0018] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0019] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0020] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0021] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0022] Continue to refer to Figure 1 The diagram illustrates a flow 100 of some embodiments of a virtual item manipulation method based on emotion tags according to the present disclosure. This virtual item manipulation method based on emotion tags includes the following steps:

[0023] Step 101: Collect real-time business data sequences for virtual items.

[0024] In some embodiments, the executor of the virtual item operation method based on sentiment tags (e.g., a computing device) can collect real-time business data sequences for virtual items via wired or wireless connections. The virtual item can be an item with financial value attributes that requires updating corresponding control strategy information. In practice, virtual items can be, but are not limited to, stocks, bonds, futures, funds, options, trust products, etc. Furthermore, virtual items can also be a combination of products with different financial value attributes. The real-time business data sequence is time-series business data specific to the virtual item. In practice, real-time business data can include, but is not limited to, opening price, closing price, highest price, and lowest price. Optionally, the real-time business data sequence can also include indicators for measuring liquidity, such as trading volume. Specifically, the executor can collect real-time business data specific to the virtual item at a fixed time granularity as the aforementioned real-time business data sequence. The fixed time granularity can be any of the following: second-level time granularity, millisecond-level time granularity, or minute-level time granularity.

[0025] It should be noted that the aforementioned wireless connection methods may include, but are not limited to, 3G / 4G / 5G connections, WiFi connections, Bluetooth connections, WiMAX connections, Zigbee connections, UWB (ultra wideband) connections, and other currently known or future wireless connection methods.

[0026] It should be noted that the aforementioned computing devices can be either hardware or software. When the computing device is hardware, it can be implemented as a distributed cluster consisting of multiple servers or terminal devices, or as a single server or a single terminal device. When the computing device is software, it can be installed on the hardware devices listed above. It can be implemented as, for example, multiple software programs or software modules used to provide distributed services, or as a single software program or software module. No specific limitations are made here.

[0027] Step 102: Perform data preprocessing on the real-time business data sequence to obtain the preprocessed business data sequence.

[0028] In some embodiments, the aforementioned execution entity may perform data preprocessing on the real-time business data sequence to obtain a preprocessed business data sequence. The preprocessed business data sequence may be a business data sequence obtained after data preprocessing.

[0029] In some optional implementations of certain embodiments, the execution entity performs data preprocessing on the real-time service data sequence to obtain a preprocessed service data sequence, which may include the following steps:

[0030] The first step is to perform the following processing steps for each real-time business data in the above real-time business data sequence:

[0031] The first sub-step is to verify the validity of the aforementioned real-time business data.

[0032] In practice, the aforementioned implementing entities can perform extreme value verification on real-time business data. For example, they can perform extreme value verification on the opening price, closing price, highest price, and lowest price of real-time business data, using the corresponding valid value range. When the value exceeds the valid value range, it indicates that the extreme value verification has passed.

[0033] The second sub-step involves cleaning the real-time business data in response to the determination that the real-time business data has passed the validity check, thereby obtaining cleaned business data.

[0034] In practice, the aforementioned implementing entities can perform data cleaning on real-time business data by methods including but not limited to: removing missing values ​​and removing duplicate data, thereby obtaining cleaned business data.

[0035] The third sub-step involves formatting the cleaned business data to obtain formatted business data.

[0036] In practice, the aforementioned executing entity can format the cleaned business data according to a preset data storage format to obtain formatted business data. For example, the cleaned business data can be converted into JSON (JavaScript Object Notation) format to obtain formatted business data.

[0037] The second step is to determine the data granularity corresponding to the above real-time business data sequence.

[0038] The aforementioned data granularity represents the data acquisition frequency corresponding to the real-time business data sequence. In practice, the data acquisition frequency can represent any of the following time granularities: second-level, millisecond-level, or minute-level. Specifically, the aforementioned data granularity can be determined by defining the acquisition time interval between adjacent real-time business data.

[0039] The third step is to determine that the above data granularity is not the preset data granularity and that the above data granularity is smaller than the preset data granularity, and then to perform data downsampling on the obtained formatted business data sequence with the preset data granularity to obtain the above preprocessed business data sequence.

[0040] Among them, the preset data granularity can represent the minute-level time granularity.

[0041] As an example, the formatted business data sequence may include: [formatted business data sequence A1, ..., formatted business data sequence A60, ..., formatted business data sequence A120, ..., formatted business data sequence A180, ..., formatted business data sequence A240]. The preprocessed business data sequence obtained by downsampling the formatted business data sequence may be [formatted business data sequence A1, formatted business data sequence A60, formatted business data sequence A120, formatted business data sequence A180, formatted business data sequence A240].

[0042] Step 103: Determine the observation window based on the preprocessed business data sequence.

[0043] In some embodiments, the aforementioned executing entity can determine an observation window based on the preprocessed business data sequence. The observation window can be a time window used for determining sentiment labels. Determining the observation window allows for further data downsampling and also enables focusing on representative business data.

[0044] In some optional implementations of certain embodiments, the execution entity determines the observation window based on the preprocessed business data sequence, which may include the following steps:

[0045] The first step is to initialize the window length corresponding to the above observation window to obtain the initial observation window.

[0046] The window length of the initial observation window is referred to as the initial window length. The window length represents the amount of preprocessed business data that can be contained within the initial observation window.

[0047] The second step involves determining the observation window based on the initial window length and the preprocessed business data sequence described above:

[0048] The first sub-step is to determine the target business data sequence.

[0049] The target business data in the target business data sequence refers to the preprocessed business data in the aforementioned preprocessed business data sequence, which is located within the initial observation window.

[0050] As an example, see Figure 2 The diagram illustrates the process of determining the target business data sequence. In this process, the execution entity may use at least one preprocessed business data within the preprocessed business data sequence 201 and located within the initial observation window 202 as the target business data sequence 203.

[0051] The second sub-step is to determine the average value of the business data based on the target business data sequence.

[0052] In practice, the aforementioned implementing entity can determine the average value of the business data based on the target business data sequence using the following formula:

[0053]

[0054] in, This represents the average of the business data. i represents the sequence number, and w... k R represents the initial observation window length. i This represents the closing price included in the i-th target business data in the target business data sequence.

[0055] The third sub-step involves determining the window feature values ​​of the initial observation window based on the target business data sequence, the average business data, the first target business data, and the second target business data.

[0056] The first target business data is the target business data corresponding to the maximum value in the target business data sequence. The second target business data is the target business data corresponding to the minimum value in the target business data sequence.

[0057] In practice, the aforementioned executing entity can determine the window feature value of the initial observation window based on the target business data sequence, the average business data, the first target business data, and the second target business data, using the following formula:

[0058]

[0059] Where, f(w) k Characteristic values ​​of the representation window. This represents the average of the business data. i represents the sequence number, and w... k R represents the initial observation window length. i R represents the closing price included in the i-th target business data in the target business data sequence. maxThe opening price is included in the primary target business data. min The second target business data includes the opening price. The initial observation window length ranges from [w0, w1]. Here, w0 represents the minimum window length, and w1 represents the maximum window length. In practice, the range of the initial observation window length varies depending on the scenario. For example, for virtual items of the general futures or securities type, the range could be [15, 30].

[0060] The fourth sub-step is to determine the initial observation window as the aforementioned observation window in response to the determination that the average value of the business data is greater than or equal to a preset threshold.

[0061] The preset threshold dynamically changes based on the business scenario corresponding to the virtual item. In practice, for scenarios with large fluctuations in business data, a larger preset threshold can be used to increase the length of the observation window, allowing for better observation of data fluctuations within the window. For scenarios with small fluctuations in business data, a smaller preset threshold can be used to decrease the length of the observation window, reducing the amount of low-variation business data included and decreasing data processing volume. Specifically, the preset threshold can be obtained through testing and simulation using historical business data. For example, the preset threshold can be set between 50 and 70 based on experience, and tests can be conducted by substituting historical business data within this range (50-70). To improve the speed of setting the preset threshold, a binary search method can be used to determine it.

[0062] The third step is to increment the window length of the initial observation window in response to the determination that the average value of the business data is less than the preset threshold. This incremented window length is then used as the initial observation window, and the observation window determination step is executed again.

[0063] In practice, the aforementioned executing entity can increment the window length of the initial observation window by 1 to obtain an observation window with the increased window length, which can then be used as the initial observation window.

[0064] Step 104: Determine the sentiment label based on the target business data sequence.

[0065] In some embodiments, the aforementioned implementing entity can determine sentiment tags based on the target business data sequence. These sentiment tags characterize the risk control tendency towards the aforementioned virtual items. In practice, the tag values ​​for sentiment tags can include: 1, 0, and -1. Specifically, a sentiment tag of "1" represents a negative sentiment tag, indicating that risk control of the virtual items is required. A sentiment tag of "0" represents a neutral sentiment tag, indicating that it is impossible to effectively determine whether risk control is necessary. A sentiment tag of "1" represents a positive sentiment tag, indicating that risk control of the virtual items needs to be relaxed.

[0066] In some optional implementations of some embodiments, the aforementioned executing entity determines the sentiment label based on the target business data sequence, including:

[0067] The first step is to determine the mean business data corresponding to each target business data in the above target business data sequence based on the above target business data.

[0068] The average business data refers to the average of the opening price, closing price, highest price, and lowest price included in the target business data.

[0069] The second step is to determine the first and second sentiment values ​​based on the obtained mean business data sequence.

[0070] In practice, the aforementioned implementing entity can determine the first sentiment value based on the obtained mean business data sequence using the following formula:

[0071]

[0072] Where mkt_d represents the first sentiment value. Specifically, since the initial observation window is the aforementioned observation window when the average value of the business data is greater than or equal to a preset threshold, W k This represents the length of the observation window. `i` represents the sequence number. `df[i]` represents the i-th mean data point in the mean data sequence. `df[i-1]` represents the (i-1)-th mean data point in the mean data sequence. `df[0:i]` represents the mean of the 0th mean data point to the i-th mean data point in the mean data sequence. `max(df[0:i])` represents the maximum mean data point from the 0th mean data point to the l-th mean data point in the mean data sequence.

[0073] In practice, the aforementioned implementing entity can determine the second sentiment value based on the obtained mean business data sequence using the following formula:

[0074]

[0075] Where mkt_vd represents the first sentiment value. Specifically, since the initial observation window is the aforementioned observation window when the average value of the business data is greater than or equal to a preset threshold, W k This represents the length of the observation window. `i` represents the sequence number. `df[i]` represents the i-th mean data point in the mean data sequence. `df[i-1]` represents the (i-1)-th mean data point in the mean data sequence. `df[0:i]` represents the mean of the 0th mean data point to the i-th mean data point in the mean data sequence. `min(df[0:i])` represents the minimum mean data point among the mean of the 0th mean data point to the i-th mean data point in the mean data sequence.

[0076] Alternatively, the aforementioned implementing entity may also determine the first sentiment value based on the obtained mean business data sequence using the following formula:

[0077]

[0078] Where mkt_d represents the first sentiment value. Specifically, since the initial observation window is the aforementioned observation window when the average value of the business data is greater than or equal to a preset threshold, W k `i` represents the length of the observation window. `i` represents the sequence number. `df[i]` represents the i-th mean data point in the mean data sequence. `df[i-1]` represents the (i-1)-th mean data point in the mean data sequence. `df[0:i]` represents the mean of the 0th mean data point to the i-th mean data point in the mean data sequence. `max(df[0:i])` represents the maximum mean data point from the 0th mean data point to the l-th mean data point in the mean data sequence. `ewa()` represents the exponentially weighted average.

[0079] Alternatively, the aforementioned implementing entity may also determine the second sentiment value based on the obtained mean business data sequence using the following formula:

[0080]

[0081] Where mkt_vd represents the first sentiment value. Specifically, since the initial observation window is the aforementioned observation window when the average value of the business data is greater than or equal to a preset threshold, W k`i` represents the length of the observation window. `i` represents the sequence number. `df[i]` represents the i-th mean data point in the mean data sequence. `df[i-1]` represents the (i-1)-th mean data point in the mean data sequence. `df[0:i]` represents the mean of the 0th mean data point to the i-th mean data point in the mean data sequence. `min(df[0:i])` represents the minimum mean data point among the mean of the 0th mean data point to the l-th mean data point in the mean data sequence. `ewa()` represents the exponentially weighted average.

[0082] The third step is to determine the minimum emotion value between the first emotion value and the second emotion value mentioned above as the target emotion value.

[0083] In practice, the process of determining the emotional value of a target can be represented by the following formula:

[0084] mkt = min(mkt_d, mkt_vd)

[0085] Where mkt represents the target sentiment value, mkt_d represents the first sentiment value, and mkt_vd represents the second sentiment value.

[0086] The fourth step is to determine the mean sentiment value, extreme sentiment value, and mean sentiment value within the preset time period.

[0087] Wherein, the aforementioned preset time period is the time period preceding the collection time corresponding to the aforementioned real-time business data sequence; the aforementioned extreme sentiment value is the minimum sentiment value within the aforementioned preset time period; and the aforementioned mean sentiment value is the average sentiment value within the aforementioned preset time period. Optionally, the mean sentiment value can also represent the median sentiment value within the preset time period. Specifically, the preset time period can be the past 30 days.

[0088] Fifth step: In response to determining that the target emotion value is less than the mean emotion value, determine the probability of the first emotion tendency based on the mean emotion value, the target emotion value, the extreme emotion value, and the second extreme emotion value.

[0089] Step 6: In response to determining that the target emotion value is greater than or equal to the mean emotion value, the preset emotion tendency probability is determined as the first emotion tendency probability.

[0090] The preset probability of emotional tendency is 0.

[0091] In practice, the probability of the first emotional tendency can be determined by the following formula:

[0092]

[0093] Where rt represents the probability of the first emotional tendency, and mkt represents the target emotional value.min This represents the sentiment value. (mkt) the This represents the average sentiment score.

[0094] Optionally, determining the sentiment label based on the target business data sequence also includes:

[0095] The first step is to determine the probability of the second emotional tendency.

[0096] In practice, the aforementioned executing entity can re-execute steps 101 to 104 at a specific time later than the collection time of the real-time business data sequence to obtain the second emotional tendency probability. The calculation method for the second emotional tendency probability is the same as the determination method for the first emotional tendency probability.

[0097] The second step is to perform weighted correction based on the above-mentioned first emotional tendency probability and the above-mentioned second emotional tendency probability to obtain the corrected emotional tendency probability.

[0098] In practice, the aforementioned implementing entity can use the above-mentioned first emotional tendency probability and the above-mentioned second emotional tendency probability to perform weighted correction using the following formula to obtain the corrected emotional tendency probability:

[0099] rt_rev=q·rt+(1-q)·rt_2

[0100] Where rt_rev represents the corrected emotional tendency probability. rt represents the first emotional tendency probability. rt_2 represents the second emotional tendency probability. q represents the weight corresponding to the first emotional tendency probability. In practice, the weight corresponding to the first emotional tendency probability ranges from 0.3 to 0.5.

[0101] The correction of the initial emotional tendency probability is achieved by determining the corrected emotional tendency probability. Furthermore, the calculation of emotional tendency probability can be performed in time periods; for example, a single trading day can be divided into three time intervals (morning, afternoon, and night) for separate calculations.

[0102] The third step is to determine the aforementioned emotion label based on either the first emotion tendency probability or the corrected emotion tendency probability:

[0103] In practice, without weighted correction, the above emotion labels can be determined based on the probability of the first emotion tendency using the following formula:

[0104]

[0105] Here, flag represents the sentiment label. rt represents the probability of the first sentiment tendency. i This indicates the last target business data (including the closing price) within the observation window. i+1This indicates the first target business data (including the closing price) after the observation window.

[0106] In practice, when performing weighted correction, the above emotional labels can be determined based on the probability of the corrected emotional tendency using the following formula:

[0107]

[0108] Here, `flag` represents the emotion label, and `rt_rev` represents the probability of the corrected emotional tendency. i This indicates the last target business data (including the closing price) within the observation window. i+1 This indicates the first target business data (including the closing price) after the observation window.

[0109] Step 105: In response to determining that the emotion label is a negative emotion label, generate negative adjustment control strategy information for the virtual item based on the current control strategy information and emotion label corresponding to the virtual item, and use it as the updated control strategy information.

[0110] In some embodiments, in response to determining that the sentiment label is a negative sentiment label, the aforementioned executing entity can generate negative adjustment control strategy information for the virtual item based on the current control strategy information corresponding to the virtual item and the sentiment label, as the updated control strategy information. Specifically, a sentiment label with a value of "-1" is considered a negative sentiment label. The current control strategy information represents the current item control strategy for the virtual item. Specifically, the current control strategy information includes, but is not limited to: current position size, margin ratio, virtual item percentage configuration, minimum position size, and maximum position size.

[0111] As an example, the aforementioned executor can generate negative adjustment control strategy information for virtual items based on preset control strategy rules, the current control strategy information corresponding to the virtual item, and the emotion tag, as the updated control strategy information. The control strategy rules are pre-built trigger rules for changes in control strategies under different scenarios.

[0112] Optionally, the aforementioned implementing entity can generate updated control strategy information based on the current control strategy information and sentiment labels using a control strategy information generation model. Specifically, the control strategy information model includes a global feature library, a control strategy information feature extraction model, and a control strategy information prediction model. The control strategy information feature extraction model and the control strategy information prediction model employ an encoder-decoder network model. The control strategy information feature extraction model is used to extract features from the current control strategy information. The control strategy information prediction model is used to generate the updated control strategy information. Specifically, when the sentiment label is negative, the risk control tendency corresponding to the generated updated control strategy information is constrained to be higher than the risk control tendency corresponding to the current control strategy information. Furthermore, to avoid getting trapped in local solutions by relying solely on the updated control strategy information from the current virtual item operation, a global feature library is introduced. This global feature library is constructed based on expert experience and stores features corresponding to real-time domain changes in the virtual item's domain. The decoding network model obtains the updated control strategy information by using the input of the control strategy information feature extraction model as a local constraint and the features corresponding to real-time domain changes in the virtual item's domain in the global feature library as global constraints. Furthermore, considering that global features can be obtained by combining multiple data sources, a multimodal model is used for feature extraction in the construction of global features. This multimodal model can include a text feature extraction module, an image feature extraction module, a video feature extraction module, and an alignment layer for aligning the output features of these modules. In practice, the text feature extraction module can adopt a Seq2Seq model structure. Additionally, since a video can be understood as a series of consecutive frames, the video feature extraction module and the image feature extraction module can share a common model structure. Specifically, a Video-LLaMa model can be used for content understanding of images and videos to output image features or video features specific to the video or image. Furthermore, considering that the features output by different modules may reside in different feature spaces, an alignment layer is used to align the features output by the text, image, and video feature extraction modules, mapping the features to the same or similar feature spaces. Through the aforementioned control strategy information generation model, given known emotion tags, corresponding control strategies can be automatically generated, especially for virtual items with complex configurations, improving the efficiency and accuracy of control strategy information generation. In practice, when there are errors in the operation of virtual items, such as when there are no abnormalities in the operation of virtual items, but the trusted device is locked or the firewall is activated incorrectly, it may affect the normal use of the device.Therefore, based on the accurately generated control strategy information, this disclosure can achieve precise and effective control of virtual items, especially for abnormal virtual item operations, such as locking trusted devices or activating firewalls.

[0113] Step 106: In response to determining that the emotion label is a positive emotion label, generate positive adjustment control strategy information for virtual items based on the current control strategy information and the emotion label, as the updated control strategy information.

[0114] In some embodiments, in response to determining that the emotion label is a positive emotion label, the aforementioned executing entity can generate positive adjustment control strategy information for virtual items based on the current control strategy information and the emotion label, as updated control strategy information.

[0115] As an example, the aforementioned executor can generate negative adjustment control strategy information for virtual items based on preset control strategy rules, the current control strategy information corresponding to the virtual item, and the emotion tag, as the updated control strategy information. The control strategy rules are pre-built trigger rules for changes in control strategies under different scenarios.

[0116] Optionally, when the emotion label is a positive emotion label, the risk control tendency corresponding to the generated updated control strategy information will be constrained to be lower than the risk control tendency corresponding to the current control strategy information.

[0117] Step 107: Adjust the virtual items according to the current control strategy information or the updated control strategy information.

[0118] In some embodiments, the aforementioned executing entity can regulate virtual items based on current or updated control strategy information. Specifically, when updated control strategy information exists, it can be prioritized for regulation. When updated control strategy information does not exist, the current control strategy information can continue to be used for regulation. Specifically, the aforementioned executing entity can adjust the current holding amount, margin ratio, and virtual item proportion of virtual items based on the current or updated control strategy information to achieve the purpose of regulating virtual items.

[0119] As an example, the aforementioned implementing entity regulates virtual items based on current or updated control policy information, including:

[0120] The first step is to determine the abnormal control information based on the current control strategy information or the updated control strategy information, as well as the emotion tags and operational anomaly prediction model.

[0121] The anomaly control information includes anomaly type and anomaly probability. Anomaly type represents the type of anomaly that may occur when executing the current or updated control strategy information. Anomaly probability represents the probability of an anomaly type occurring. The operation anomaly prediction model includes a control strategy information feature extraction model and an anomaly prediction layer. In practice, the encoding network included in the virtual item operation model can be reused to extract features from the current or updated control strategy information to generate control information features, thereby reducing the model size. Then, the executing entity can input the control information features into the anomaly prediction layer to obtain the anomaly type and the corresponding anomaly probability. In practice, the anomaly prediction layer can be implemented using a fully connected layer. Specifically, during the training phase, the anomaly prediction model is trained using supervised training. The training samples are historical control strategy information labeled with corresponding anomaly types, and the corresponding training labels are the anomaly types corresponding to the historical control strategy information.

[0122] The second step is to determine the control instructions corresponding to the abnormal control information in the abnormal control decision tree when it is determined that the above-mentioned abnormal control information indicates that there is a control abnormality in the current control strategy information or the updated control strategy information.

[0123] The anomaly control decision tree records different anomaly types and their corresponding control instructions. Control instructions represent the anomaly control-related commands for virtual items.

[0124] The third step is to lock the authorized devices related to the virtual items according to the above control instructions, or activate the firewall to block operations related to the virtual items.

[0125] As an example, a control command could be a virtual item data locking command, which locks the business data related to virtual items to prevent the data from being tampered with.

[0126] As another example, the control command could be a lock on a trusted device associated with a virtual item. This trusted device could be a trusted mobile terminal. By locking the trusted device, operations related to the virtual item cannot be performed through it, thus preventing abnormal operations and potential data anomalies.

[0127] As another example, control commands can be data leakage prevention commands targeting data storage devices related to virtual items. For instance, when unauthorized or highly unusual data modifications or deletions occur on business data related to virtual items, the firewall can be activated to block such operations, thereby preventing abnormal actions.

[0128] Optionally, when the abnormal control information indicates that there is no control abnormality in the current control strategy information or the updated control strategy information, the current control strategy information or the updated control strategy information is executed to regulate the virtual items.

[0129] In some optional implementations of certain embodiments, the execution entity regulates virtual items based on current or updated control policy information, which may include the following steps:

[0130] The first step is to generate adjustment prompts for the virtual items based on the current control strategy information or the updated control strategy information.

[0131] In practice, when it is necessary to adjust virtual items, the relevant personnel need to be notified for confirmation. Therefore, adjustment prompts related to virtual items can be generated.

[0132] The second step is to send the aforementioned control prompts to the target terminal.

[0133] The target terminal mentioned above is the terminal bound to the processing account associated with the aforementioned virtual item control. In practice, the control notification information can be transmitted using encrypted methods.

[0134] The third step is to respond to the control confirmation command initiated by the target terminal and execute the current control strategy information or the updated control strategy information.

[0135] In practice, by adding a manual review and confirmation step before execution, potential control errors that may occur during automatic execution can be avoided, thus increasing the security of the control process.

[0136] The fourth step is to respond to the control update instruction initiated by the target terminal and update the current control strategy information or the updated control strategy information according to the control update instruction to obtain the control strategy information after the second update or the control strategy information after the third update.

[0137] In practice, the adjustment update command can be initiated by the processing object associated with the aforementioned virtual item adjustment, and can be an adjustment item targeting the current control policy information or the updated control policy information. Specifically, when the adjustment update command targets the current control policy information, the current control policy information can be updated based on the adjustment update quality to obtain the second-updated control policy information. When the adjustment update command targets the updated control policy information, the updated control policy information can be updated based on the adjustment update quality to obtain the third-updated control policy information.

[0138] The fourth step is to adjust the virtual items according to the control strategy information after the second or third update.

[0139] In practice, the aforementioned implementing entities can execute the control strategy information after the second or third update to regulate the aforementioned virtual items.

[0140] The above embodiments of this disclosure have the following beneficial effects: The virtual item operation method based on emotion tags in some embodiments of this disclosure effectively prevents and protects against abnormal operations, avoiding potential data risks and ensuring device security. Specifically, the reason for the above problems is that, for frequent virtual item operations, using methods such as manual anomaly control makes it difficult to comprehensively and effectively detect abnormal virtual item operations, thus making it impossible to accurately and effectively control and schedule virtual items. Especially when abnormal virtual item operations exist, it is impossible to effectively prevent and protect against abnormal operations, which may lead to data risks and device security risks. Based on this, the virtual item operation method based on emotion tags in some embodiments of this disclosure first collects real-time business data sequences for virtual items to obtain real-time data for virtual items. Second, the real-time business data sequences are preprocessed to obtain preprocessed business data sequences. Data preprocessing improves data quality and eliminates noise that low-quality data may cause to subsequent results. Then, an observation window is determined based on the preprocessed business data sequences. In practice, an observation window is dynamically determined by combining the preprocessed business data sequence to facilitate the subsequent capture of abnormal control tendencies. Further, a sentiment label is determined based on the target business data sequence, where the target business data sequence is the preprocessed business data within the observation window from the aforementioned preprocessed business data sequence, and the sentiment label represents the risk control tendency towards the aforementioned virtual item. This, combined with the preprocessed business data within the observation window, quantifies the risk control tendency towards the virtual item through the sentiment label. Furthermore, in response to determining that the sentiment label is a negative sentiment label, negative adjustment control strategy information for the aforementioned virtual item is generated based on the current control strategy information corresponding to the virtual item and the sentiment label, serving as the updated control strategy information. Subsequently, in response to determining that the sentiment label is a positive sentiment label, positive adjustment control strategy information for the aforementioned virtual item is generated based on the current control strategy information and the sentiment label, serving as the updated control strategy information. Based on this, the current control strategy information and emotion tags are combined to update the corresponding control strategy information. Finally, according to the current control strategy information or the updated control strategy information, the virtual item is regulated, including: determining abnormal control information based on the current control strategy information or the updated control strategy information, as well as the emotion tags and the operation anomaly prediction model; in response to determining that the abnormal control information indicates that there is a control anomaly in the current control strategy information or the updated control strategy information, determining the control instruction corresponding to the abnormal control information in the abnormal control decision tree; and locking the authorized device related to the virtual item or activating the firewall to block the operation related to the virtual item according to the control instruction.In particular, by locking the trusted device or activating the firewall to block abnormal virtual item operations, abnormal operations can be effectively prevented and protected, avoiding potential data risks and ensuring device security.

[0141] Further reference Figure 3 As an implementation of the methods shown in the above figures, this disclosure provides some embodiments of a virtual item manipulation device based on emotion tags. These device embodiments are similar to... Figure 1 Corresponding to the method embodiments shown, this emotion tag-based virtual item manipulation device can be specifically applied to various electronic devices.

[0142] like Figure 3As shown, some embodiments of the virtual item operation device 300 based on emotion tags include: a collection unit 301, a data preprocessing unit 302, a first determination unit 303, a second determination unit 304, a first generation unit 305, a second generation unit 306, and a virtual item scheduling unit 307. The acquisition unit 301 is configured to acquire real-time business data sequences for virtual items; the data preprocessing unit 302 is configured to preprocess the real-time business data sequences to obtain preprocessed business data sequences; the first determining unit 303 is configured to determine an observation window based on the preprocessed business data sequences; the second determining unit 304 is configured to determine a sentiment label based on a target business data sequence, wherein the target business data sequence is the preprocessed business data within the observation window in the preprocessed business data sequence, and the sentiment label represents the risk control tendency for the virtual items; the first generating unit 305 is configured to, in response to determining that the sentiment label is a negative sentiment label, generate negative adjustment control strategy information for the virtual items based on the current control strategy information corresponding to the virtual items and the sentiment label, as updated control strategy information; the second generating unit 305 is configured to... Unit 306 is configured to, in response to determining that the aforementioned emotion tag is a positive emotion tag, generate positive adjustment control strategy information for the aforementioned virtual item based on the aforementioned current control strategy information and the aforementioned emotion tag, as updated control strategy information; Virtual item control unit 307 is configured to perform virtual item control on the aforementioned virtual item based on the aforementioned current control strategy information or the aforementioned updated control strategy information, including: determining abnormal control information based on the current control strategy information or the updated control strategy information, as well as the emotion tag and the operation anomaly prediction model; in response to determining that the abnormal control information indicates that the current control strategy information or the updated control strategy information has a control anomaly, determining the control instruction corresponding to the abnormal control information in the abnormal control decision tree; and locking the authorized device related to the virtual item or activating the firewall to block the virtual item-related operations based on the control instruction.

[0143] It is understandable that the units described in the emotion-tag-based virtual item manipulation device 300 are related to the reference... Figure 1 The steps in the described method correspond accordingly. Therefore, the operations, features, and beneficial effects described above for the method also apply to the emotion-tag-based virtual item manipulation device 300 and the units contained therein, and will not be repeated here.

[0144] The following is for reference. Figure 4 It shows a schematic diagram of the structure of an electronic device (e.g., a computing device) 400 suitable for implementing some embodiments of the present disclosure. Figure 4The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of this disclosure.

[0145] like Figure 4 As shown, the electronic device 400 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 401, which can perform various appropriate actions and processes according to a program stored in a read-only memory 402 or a program loaded from a storage device 408 into a random access memory 403. The random access memory 403 also stores various programs and data required for the operation of the electronic device 400. The processing unit 401, the read-only memory 402, and the random access memory 403 are interconnected via a bus 404. An input / output interface 405 is also connected to the bus 404.

[0146] Typically, the following devices can be connected to the input / output interface 405: input devices 406 including, for example, a touchscreen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 407 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 408 including, for example, magnetic tape, hard disk, etc.; and communication devices 409. Communication device 409 allows electronic device 400 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 4 An electronic device 400 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively. Figure 4 Each box shown can represent a device or multiple devices as needed.

[0147] In particular, according to some embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 409, or installed from a storage device 408, or installed from a read-only memory 402. When the computer program is executed by the processing device 401, it performs the functions defined in the methods of some embodiments of this disclosure.

[0148] It should be noted that, in some embodiments of this disclosure, the computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In some embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0149] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0150] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to: acquire a real-time business data sequence for the virtual item; preprocess the real-time business data sequence to obtain a preprocessed business data sequence; determine an observation window based on the preprocessed business data sequence; determine a sentiment label based on a target business data sequence, wherein the target business data sequence is the preprocessed business data within the observation window in the preprocessed business data sequence, and the sentiment label represents the risk control tendency for the virtual item; in response to determining that the sentiment label is a negative sentiment label, generate negative adjustment control strategy information for the virtual item based on the current control strategy information corresponding to the virtual item and the sentiment label, as updated control strategy information. The system takes the following steps: In response to determining that the aforementioned emotion tag is a positive emotion tag, it generates positive adjustment control strategy information for the aforementioned virtual item based on the aforementioned current control strategy information and the aforementioned emotion tag, as updated control strategy information; it then regulates the aforementioned virtual item based on the aforementioned current control strategy information or the aforementioned updated control strategy information, including: determining abnormal control information based on the aforementioned current control strategy information or the aforementioned updated control strategy information, as well as the emotion tag and the operation anomaly prediction model; in response to determining that the abnormal control information indicates a control anomaly in the current control strategy information or the updated control strategy information, it determines the control instruction corresponding to the abnormal control information in the abnormal control decision tree; and based on the control instruction, it locks the authorized device related to the virtual item or activates the firewall to block operations related to the virtual item.

[0151] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0152] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0153] The units described in some embodiments of this disclosure can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor may be described as including a data acquisition unit, a data preprocessing unit, a first determination unit, a second determination unit, a first generation unit, a second generation unit, and a virtual item scheduling unit. The names of these units do not necessarily limit the specific unit itself. For example, the second generation unit may also be described as "a unit that, in response to determining that the aforementioned emotion tag is a positive emotion tag, generates positive adjustment control strategy information for the aforementioned virtual item based on the aforementioned current control strategy information and the aforementioned emotion tag, as updated control strategy information."

[0154] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0155] The above description is merely a selection of preferred embodiments of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A method for manipulating virtual items based on emotion tags, applied to trust devices or firewalls, comprising: Collect real-time business data sequences for virtual items; The real-time business data sequence is preprocessed to obtain a preprocessed business data sequence; The observation window is determined based on the preprocessed business data sequence; Based on the target business data sequence, a sentiment label is determined, wherein the target business data sequence is the preprocessed business data located within the observation window in the preprocessed business data sequence, and the sentiment label represents the risk control tendency for the virtual item; In response to determining that the emotion tag is a negative emotion tag, negative adjustment control strategy information for the virtual item is generated based on the current control strategy information corresponding to the virtual item and the emotion tag, and used as the updated control strategy information; In response to determining that the emotion tag is a positive emotion tag, positive adjustment control strategy information for the virtual item is generated based on the current control strategy information and the emotion tag, as the updated control strategy information; Based on the current control policy information or the updated control policy information, the virtual item is adjusted, including: Based on the current control strategy information or the updated control strategy information, as well as the emotion label and the operation anomaly prediction model, determine the abnormal control information; In response to determining that the abnormal control information indicates that there is a control abnormality in the current control strategy information or the updated control strategy information, the control instruction corresponding to the abnormal control information in the abnormal control decision tree is determined. According to the control command, the trusted device related to the virtual item is locked, or the firewall is activated to block operations related to the virtual item.

2. The method according to claim 1, wherein, The step of preprocessing the real-time service data sequence to obtain a preprocessed service data sequence includes: For each real-time service data in the real-time service data sequence, perform the following processing steps: Perform data validity verification on the real-time business data; In response to determining that the real-time business data has passed the validity verification, the real-time business data is cleaned to obtain cleaned business data. The cleaned business data is formatted to obtain formatted business data; Determine the data granularity corresponding to the real-time business data sequence, wherein the data granularity characterizes the data acquisition frequency corresponding to the real-time business data sequence; In response to determining that the data granularity is not a preset data granularity and that the data granularity is smaller than the preset data granularity, the obtained formatted business data sequence is downsampled using the preset data granularity to obtain the preprocessed business data sequence.

3. The method according to claim 2, wherein, The step of determining the observation window based on the preprocessed business data sequence includes: Initialize the window length corresponding to the observation window to obtain the initial observation window, wherein the window length of the initial observation window is the initial window length; Based on the initial window length and the preprocessed business data sequence, the following observation window determination steps are performed: Determine the target business data sequence, wherein the target business data in the target business data sequence is the preprocessed business data located within the initial observation window in the preprocessed business data sequence; Determine the average value of the business data based on the target business data sequence; Based on the target business data sequence, the business data mean, the first target business data, and the second target business data, the window feature values ​​of the initial observation window are determined, wherein the first target business data is the target business data corresponding to the maximum value in the target business data sequence, and the second target business data is the target business data corresponding to the minimum value in the target business data sequence. In response to determining that the average value of business data is greater than or equal to a preset threshold, an initial observation window is determined as the observation window, wherein the preset threshold changes dynamically according to the business scenario corresponding to the virtual item; In response to the determination that the average value of the business data is less than the preset threshold, the window length of the initial observation window is incremented to obtain an observation window with an increased window length, which is then used as the initial observation window. The observation window determination step is then executed again.

4. The method according to claim 3, wherein, The step of determining the sentiment label based on the target business data sequence includes: For each target business data in the target business data sequence, the mean business data corresponding to the target business data is determined based on the target business data. Based on the obtained mean business data sequence, the first sentiment value and the second sentiment value are determined respectively; The minimum emotion value between the first emotion value and the second emotion value is determined as the target emotion value; Determine the mean sentiment value, the first extreme sentiment value, and the second extreme sentiment value within a preset time period, wherein the preset time period is the time period preceding the collection time corresponding to the real-time business data sequence, the first extreme sentiment value is the minimum value of the sentiment value within the preset time period, and the mean sentiment value is the average value of the sentiment value within the preset time period. In response to determining that the target emotion value is less than the mean emotion value, a first emotion tendency probability is determined based on the mean emotion value, the target emotion value, the first extreme emotion value, and the second extreme emotion value; In response to determining that the target emotion value is greater than or equal to the mean emotion value, a preset emotion tendency probability is determined as the first emotion tendency probability, wherein the preset emotion tendency probability is 0.

5. The method according to claim 4, wherein, The step of determining the sentiment label based on the target business data sequence also includes: Determine the probability of the second emotional tendency; The first emotional tendency probability and the second emotional tendency probability are weighted and corrected to obtain the corrected emotional tendency probability. The emotion label is determined based on the first emotion tendency probability or the corrected emotion tendency probability.

6. The method according to claim 5, wherein, The step of adjusting the virtual items according to the current control policy information or the updated control policy information includes: Based on the current control strategy information or the updated control strategy information, generate adjustment prompt information for the virtual item; The control prompt information is sent to the target terminal, wherein the target terminal is a terminal bound to the processing object account associated with the control of the virtual item; In response to receiving a control confirmation command initiated by the target terminal, the current control strategy information or the updated control strategy information is executed. In response to receiving a control update instruction initiated by the target terminal, the current control strategy information or the updated control strategy information is updated according to the control update instruction to obtain the control strategy information after the second update or the control strategy information after the third update. Based on the control strategy information after the second update or the control strategy information after the third update, the virtual items are regulated.

7. A virtual item manipulation device based on emotion tags, applied to a trust device or firewall, comprising: The acquisition unit is configured to acquire real-time business data sequences for virtual items; A data preprocessing unit is configured to preprocess the real-time business data sequence to obtain a preprocessed business data sequence. The first determining unit is configured to determine an observation window based on the preprocessed business data sequence; The second determining unit is configured to determine the sentiment label of the target business data sequence, wherein the target business data sequence is the preprocessed business data located within the observation window in the preprocessed business data sequence, and the sentiment label represents the risk control tendency for the virtual item; The first generation unit is configured to, in response to determining that the emotion tag is a negative emotion tag, generate negative adjustment control strategy information for the virtual item based on the current control strategy information corresponding to the virtual item and the emotion tag, as the updated control strategy information; The second generation unit is configured to, in response to determining that the emotion tag is a positive emotion tag, generate positive adjustment control strategy information for the virtual item based on the current control strategy information and the emotion tag, as the updated control strategy information; A virtual item control unit is configured to control the virtual item based on the current control policy information or the updated control policy information, including: Based on the current control strategy information or the updated control strategy information, as well as the emotion label and the operation anomaly prediction model, determine the abnormal control information; In response to determining that the abnormal control information indicates that there is a control abnormality in the current control strategy information or the updated control strategy information, the control instruction corresponding to the abnormal control information in the abnormal control decision tree is determined. According to the control command, the trusted device related to the virtual item is locked, or the firewall is activated to block operations related to the virtual item.

8. An electronic device, comprising: One or more processors; A storage device on which one or more programs are stored; When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1 to 6.

9. A computer-readable medium having a computer program stored thereon, wherein, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 6.

Images