Anchor user data intelligent processing method, system, device and medium

Abstract

The invention discloses an intelligent anchor user data processing method, system and device and a medium. The method comprises the steps that a performance assessment score strategy is generated according to a performance target setting instruction; collecting whole live broadcast data in real time; obtaining anchor time period assessment data corresponding to each anchor through the whole live broadcast data, an anchor actual live broadcast time period set obtained based on anchor user information, the whole live broadcast data and a face recognition model, and a forbidden word time point set obtained based on forbidden word information, the whole live broadcast data and a voice recognition model; if an anchor performance assessment instruction is received, anchor time period assessment data corresponding to each anchor is obtained according to an assessment period, and an anchor performance assessment data set is obtained; and obtaining a performance assessment result corresponding to each anchor according to the performance assessment score strategy and the anchor performance assessment data set. According to the invention, the live broadcast process data of each anchor can be quickly, accurately and automatically acquired and processed, and the data processing efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of live broadcast technology, and in particular to a method, system, computer equipment and storage medium for intelligent processing of anchor user data. Background Art

[0002] With the rapid development of Internet technology and the popularization of e-commerce, e-commerce live streaming, as an emerging business model, is gradually becoming an important force to promote consumption growth. Among them, anchors have become an important role in the e-commerce industry, and it is usually necessary to analyze the anchor's live broadcast process data to obtain the anchor's assessment results, and then urge the anchor to improve his skills and improve the quality of live broadcasts. However, project managers generally need to correspond to multiple projects, and each project has anchors who take turns to broadcast live, which makes it difficult to analyze the actual live broadcast process data of each anchor, such as the actual live broadcast duration and live broadcast effect, and it is impossible to automatically and quickly obtain the actual live broadcast process data corresponding to each anchor for analysis. Summary of the invention

[0003] The embodiments of the present invention provide a method, system, computer device and storage medium for intelligent processing of anchor user data, aiming to solve the problem in the prior art that it is impossible to automatically and quickly obtain the actual live broadcast process data corresponding to each anchor for analysis.

[0004] In a first aspect, an embodiment of the present invention provides a method for intelligently processing anchor user data, which includes:

[0005] In response to an information entry instruction, obtaining banned word information and anchor user information corresponding to the information entry instruction for storage;

[0006] In response to a performance target setting instruction, generating a corresponding performance appraisal scoring strategy according to the performance target setting instruction;

[0007] If a live broadcast start instruction is received, live broadcast data corresponding to the live broadcast start instruction is collected in real time to obtain the whole live broadcast data;

[0008] Based on the anchor user information, the entire live broadcast data and a preset face recognition model, the actual live broadcast time period corresponding to each anchor in the entire live broadcast data is obtained, and the actual live broadcast time period set of the anchor is obtained by combining;

[0009] Based on the banned word information, the entire live broadcast data and a preset speech recognition model, the time points at which banned words appear in the entire live broadcast data are obtained, and a banned word time point set is obtained by combining them;

[0010] The anchor time period assessment data corresponding to each anchor in the entire live broadcast data is obtained by integrating the entire live broadcast data, the anchor's actual live broadcast time period set and the banned word time point set;

[0011] If an anchor performance assessment instruction is received, the anchor time period assessment data corresponding to each anchor in the anchor user information is obtained according to a preset assessment period, and the anchor performance assessment data set is obtained by combining them;

[0012] The performance appraisal result corresponding to each anchor in the anchor user information is obtained according to the performance appraisal scoring strategy and the anchor performance appraisal data set.

[0013] In a second aspect, an embodiment of the present invention provides a system for intelligently processing anchor user data, which includes:

[0014] An information management unit, configured to, in response to an information entry instruction, obtain and store banned word information and anchor user information corresponding to the information entry instruction;

[0015] A performance management unit, configured to respond to a performance target setting instruction and generate a corresponding performance appraisal scoring strategy according to the performance target setting instruction;

[0016] A data collection unit, for collecting live broadcast data corresponding to the live broadcast start instruction in real time if a live broadcast start instruction is received, so as to obtain the whole live broadcast data;

[0017] The anchor analysis unit is used to obtain the anchor's actual live broadcast time period corresponding to each anchor in the entire live broadcast data based on the anchor user information, the entire live broadcast data and a preset face recognition model, and combine them to obtain an anchor's actual live broadcast time period set;

[0018] A banned word analysis unit, configured to obtain the time points at which banned words appear in the entire live broadcast data based on the banned word information, the entire live broadcast data, and a preset speech recognition model, and combine them to obtain a banned word time point set;

[0019] A first data integration unit is used to integrate the entire live broadcast data, the actual live broadcast time period set of the anchor and the banned word time point set to obtain the anchor time period assessment data corresponding to each anchor in the entire live broadcast data;

[0020] The second data integration unit is used for obtaining the anchor time period assessment data corresponding to each anchor in the anchor user information according to a preset assessment period if an anchor performance assessment instruction is received, and combining them to obtain an anchor performance assessment data set;

[0021] The performance evaluation unit is used to obtain the performance evaluation result corresponding to each anchor in the anchor user information according to the performance evaluation scoring strategy and the anchor performance evaluation data set.

[0022] In a third aspect, an embodiment of the present invention provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for intelligently processing anchor user data of the first aspect is implemented.

[0023] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the method for intelligent processing of anchor user data of the first aspect mentioned above.

[0024] The embodiment of the present invention provides a method, system, device and medium for intelligent processing of anchor user data, the method includes generating a performance assessment scoring strategy according to a performance target setting instruction; collecting the whole live broadcast data in real time; obtaining the anchor time period assessment data corresponding to each anchor through the whole live broadcast data, the anchor user information, the whole live broadcast data and the anchor actual live broadcast time period set obtained by the face recognition model, and the banned word time point set obtained based on the banned word information, the whole live broadcast data and the voice recognition model; if the anchor performance assessment instruction is received, the anchor time period assessment data corresponding to each anchor is obtained according to the assessment cycle to obtain the anchor performance assessment data set; and the performance assessment result corresponding to each anchor is obtained according to the performance assessment scoring strategy and the anchor performance assessment data set. The present invention can quickly and accurately automatically acquire and process the live broadcast process data of each anchor, thereby improving the data processing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0026] Figure 1 A schematic diagram of a flow chart of a method for intelligently processing anchor user data provided by an embodiment of the present invention;

[0027] Figure 2 A schematic block diagram of a system for intelligently processing anchor user data provided by an embodiment of the present invention;

[0028] Figure 3 A schematic block diagram of a computer device provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0029] 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 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.

[0030] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0031] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0032] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0033] See also Figure 1 , Figure 1 The present invention is a flowchart of a method for intelligently processing anchor user data provided by an embodiment of the present invention. The method for intelligently processing anchor user data provided by the embodiment of the present invention includes steps S11 to S18, wherein the anchor user data may specifically be anchor performance evaluation data. The method for intelligently processing anchor user data provided by the embodiment of the present invention can realize performance evaluation of anchors.

[0034] S11. In response to an information entry instruction, obtain banned word information and anchor user information corresponding to the information entry instruction and store them.

[0035] In this embodiment, the information entry instruction is triggered by the relevant management personnel (such as the project manager), and the banned word information and anchor user information set by the relevant management personnel are stored through the information entry instruction. Among them, the information entry instruction includes a first information entry instruction and a second information entry instruction, the first information entry instruction is used to enter the banned word information, and the second information entry instruction is used to enter the anchor user information. There are some banned words on each live broadcast platform, such as some words suspected of deception, stimulating consumption, false propaganda, etc. The relevant management personnel set the banned words that may be risky for each live broadcast platform of the project based on their own experience, so as to input the stored banned word information according to all platform settings of the project live broadcast, so that the banned word information can be used as the reference information required for the anchor performance appraisal later. Preferably, the banned word information includes platform information, and a banned word list corresponding to the platform information, and the platform information includes the platform type and the platform live broadcast address.

[0036] Preferably, the anchor user information includes anchor portrait, anchor level (e.g., S / A / B / C level), cooperation type (e.g., full-time / part-time), anchor identity and other related basic information. The relevant management personnel will enter the anchor user information corresponding to the recruited anchors to facilitate the subsequent performance appraisal of the recruited anchors. In addition, the relevant management personnel can also configure the corresponding live broadcast personnel for each live broadcast platform according to the anchor user information. The live broadcast personnel include anchors and operation-related personnel, and complete the preliminary scheduling work.

[0037] S12. In response to a performance target setting instruction, generate a corresponding performance appraisal scoring strategy according to the performance target setting instruction.

[0038] In this embodiment, the performance target setting instruction is triggered by the relevant management personnel (such as the project manager), and the relevant management personnel trigger the performance target setting instruction at the target setting time (such as the beginning of each month) to generate the corresponding performance appraisal scoring strategy.

[0039] In one embodiment, generating a corresponding performance appraisal scoring strategy according to the performance target setting instruction includes:

[0040] Acquire assessment indicator information corresponding to the performance target setting instruction; wherein the assessment indicator information includes deduction item information, qualitative indicator information, qualitative indicator weight, quantitative indicator information and quantitative indicator weight;

[0041] Generate a deduction strategy according to the deduction item information;

[0042] Generating a qualitative indicator scoring strategy according to the qualitative indicator information and the qualitative indicator weight;

[0043] Generate a quantitative indicator scoring strategy according to the quantitative indicator information and the quantitative indicator weight;

[0044] The score reduction strategy, the qualitative indicator scoring strategy and the quantitative indicator scoring strategy are combined to obtain a performance appraisal scoring strategy.

[0045] In this embodiment, the performance appraisal scoring strategy includes a deduction strategy, a qualitative indicator scoring strategy, and a quantitative indicator scoring strategy. Specifically, the appraisal indicator information is used to indicate the anchor performance appraisal target set by the relevant management personnel based on the anchor user information, and the appraisal indicator information includes deduction item information, qualitative indicator information, qualitative indicator weights, quantitative indicator information, and quantitative indicator weights. Further, a deduction strategy is generated based on the deduction item information, and the deduction strategy is used to obtain the anchor's deduction item results. A qualitative indicator scoring strategy is generated based on the qualitative indicator information and the qualitative indicator weights, and the qualitative indicator scoring strategy is used to obtain the anchor's qualitative indicator results. A quantitative indicator scoring strategy is generated based on the quantitative indicator information and the quantitative indicator weights, and the quantitative indicator scoring strategy is used to obtain the anchor's quantitative indicator results.

[0046] S13. If a live broadcast start instruction is received, live broadcast data corresponding to the live broadcast start instruction is collected in real time to obtain the entire live broadcast data.

[0047] In this embodiment, when the live broadcast room starts to broadcast live, the live broadcast start instruction is triggered by relevant personnel, so as to start real-time collection of the live broadcast data of the live broadcast room during the live broadcast process until the live broadcast ends, and the whole live broadcast data is obtained. Among them, the whole live broadcast data includes the whole live broadcast video data, the whole live broadcast trend data and the whole live broadcast order data. Specifically, the live broadcast video of the live broadcast room can be downloaded in real time through a preset audio and video conversion tool (such as FFMPEG tool), and the live broadcast video can be archived according to the live broadcast room and live broadcast time after the live broadcast ends, so as to obtain the whole live broadcast video data. During the live broadcast process of the live broadcast room, the number of people entering the live broadcast room, the number of people leaving, the number of people joining the group, the average viewing time per person and other data can be obtained through the system background to obtain the whole live broadcast trend data. The system background corresponds to the order management platform of each live broadcast platform, and the live broadcast order data of the live broadcast room during the live broadcast process can be obtained through SDK (Software Development Kit, Software Development Kit), such as order volume, sales volume, sales, return rate and other data, so as to obtain the whole live broadcast order data.

[0048] S14. Based on the anchor user information, the entire live broadcast data and a preset face recognition model, the actual live broadcast time period of each anchor corresponding to the entire live broadcast data is obtained, and the actual live broadcast time period set of the anchor is obtained by combining.

[0049] In this embodiment, the entire live broadcast data is used as the data source, combined with the stored anchor user information, and face recognition is performed through a face recognition model to identify the actual working time period of each anchor during the entire live broadcast, that is, the actual live broadcast time period corresponding to each anchor in the entire live broadcast data is obtained. The actual live broadcast time period represents the actual working time period of the anchor, and the actual live broadcast time period corresponding to each anchor in the entire live broadcast data is combined into an anchor actual live broadcast time period set. Since the broadcast time is a flexible working mode, the actual broadcast time will be determined within the shift according to the actual situation, so obtaining the scheduled working time period of the anchor in the scheduling work as the actual working time period may not be accurate. Therefore, the actual live broadcast time period corresponding to each anchor in the entire live broadcast data can be obtained more accurately through the anchor user information, the entire live broadcast data and the face recognition model.

[0050] In one embodiment, the acquiring the actual live broadcast time period corresponding to each anchor in the entire live broadcast data based on the anchor user information, the entire live broadcast data and a preset face recognition model includes:

[0051] Acquire the entire live video data from the entire live data, and acquire the entire live time period corresponding to the entire live video data;

[0052] Obtaining a detection point set according to the entire live broadcast period and a preset detection cycle;

[0053] For each detection point in the detection point set, a live broadcast picture corresponding to the detection point in the entire live broadcast video data is obtained, and an anchor identity corresponding to the detection point is obtained according to the face recognition model, the live broadcast picture and the anchor user information;

[0054] According to the anchor identity corresponding to each detection point in the detection point set, the actual live broadcast time period corresponding to each anchor in the entire live broadcast data is obtained.

[0055] In this embodiment, the entire live video data and the corresponding entire live time period in the entire live data are obtained. Face recognition is performed on the live screen of the entire live video data according to a preset detection cycle to confirm the identity of the anchor corresponding to the live screen of the detection point, and then the actual live time period corresponding to each anchor is obtained according to the anchor identity corresponding to each detection point.

[0056] Specifically, each detection point corresponding to the entire live video data can be determined through the entire live broadcast period and the preset detection cycle, and each detection point corresponding to the entire live video data can be combined to obtain a detection point set. For example, the entire live broadcast period is 10:00-18:00, and the detection cycle is 5s. Then, a detection is performed every 5s in the entire live broadcast period, and a total of 8760 detection points can be determined, and the corresponding detection point set is composed of these 8760 detection points. For each detection point in the detection point set, the live screen corresponding to the detection point in the entire live video data is obtained, and the live screen is parsed by the face recognition model and matched with the anchor avatar in the anchor user information, and then the anchor identity corresponding to the live screen of the detection point can be confirmed. After confirming the anchor identity corresponding to each detection point in the detection point set, the actual live broadcast period corresponding to each anchor in the entire live data can be obtained according to the anchor identity corresponding to each detection point.

[0057] In one embodiment, obtaining the actual live broadcast time period corresponding to each anchor in the entire live broadcast data according to the anchor identity corresponding to each detection point in the detection point set includes:

[0058] A determination period set is obtained according to the entire live broadcast period and a preset determination cycle; wherein the determination period set includes a plurality of determination periods, and each determination period includes a plurality of detection points;

[0059] For each determination period in the determination period set, if it is determined that all detection points contained in the determination period correspond to the same anchor identity, then the same anchor identity corresponding to all detection points contained in the determination period is determined as the target anchor identity corresponding to the determination period;

[0060] If it is determined that the detection points contained in the determination period correspond to multiple anchor identities, the target anchor identity corresponding to the determination period is determined according to the broadcasting ratio of each anchor identity in the determination period;

[0061] The actual live broadcast time period corresponding to each anchor in the entire live broadcast data is obtained according to the target anchor identity corresponding to each determination time period in the determination time period set.

[0062] In this embodiment, after detecting the anchor identity corresponding to each detection point in the detection point set, the entire live broadcast period is divided according to the preset judgment period, so that the entire live broadcast period is divided into multiple judgment periods, and the multiple judgment periods form a judgment period set. Among them, the judgment period is greater than the detection period, so that each judgment period contains multiple detection points. For example, the detection period is 5s, the judgment period is 5 minutes, and the entire live broadcast period is divided with 5 minutes as a judgment point, then the judgment period contains 60 detection points.

[0063] Secondly, determine the anchor identity to which each determination period in the determination period set belongs. Specifically, for each determination period in the determination period set, if the detection points contained in the determination period all correspond to the same anchor identity, then the same anchor identity corresponding to the detection points contained in the determination period is determined as the target anchor identity corresponding to the determination period, that is, the determination period belongs to the same anchor identity, wherein, if the same anchor identity is an empty set, the anchor identity does not contain the anchor's identity information, indicating that no anchor user information is identified at the detection points contained in the determination period, and there is no anchor in the determination period; if the same anchor identity is a non-empty set, the anchor identity contains the anchor's identity information, indicating that the same anchor user information is identified at the detection points contained in the determination period, and the same anchor is broadcasting live in the determination period. If it is determined that the detection points contained in the judgment period correspond to multiple anchor identities, it means that there are multiple anchors broadcasting live in the judgment period, then the broadcaster appearance ratio of each anchor identity in the judgment period is obtained, and the anchor identity corresponding to the largest broadcaster appearance ratio is determined as the target anchor identity corresponding to the judgment period, that is, the judgment period is attributed to the anchor with the most appearances in this period. Finally, statistics are performed based on the target anchor identity corresponding to each judgment period in the determined judgment period set to obtain the actual broadcast period corresponding to each anchor in the entire live broadcast data.

[0064] S15. Based on the banned word information, the entire live broadcast data and a preset speech recognition model, the time points at which banned words appear in the entire live broadcast data are obtained, and the banned word time point set is obtained by combining them.

[0065] In this embodiment, the entire live broadcast data is used as the data source, combined with the stored banned word information, and speech recognition is performed through a speech recognition model to obtain the time points at which banned words appear in the entire live broadcast data, and the time points at which banned words appear in the entire live broadcast data are combined into a banned word time point set to facilitate subsequent statistics of the number of times banned words appear in each anchor's live broadcast.

[0066] In one embodiment, the step of obtaining the time point at which the banned word appears in the entire live broadcast data based on the banned word information, the entire live broadcast data, and a preset speech recognition model includes:

[0067] Using the speech recognition model to perform speech recognition on the entire live broadcast data to obtain dialogue text and text time;

[0068] The banned word information, the dialogue text and the text time are matched to obtain the time point when the banned word appears in the entire live broadcast data.

[0069] In this embodiment, the audio file of the entire live broadcast data is parsed through a speech recognition model to obtain the corresponding dialogue text and text time after conversion, and the dialogue text, text time and the stored banned word information are matched to obtain the time point when the banned word appears in the dialogue text, that is, the time point when the banned word appears in the entire live broadcast data.

[0070] S16. Integrate the entire live broadcast data, the actual live broadcast time period set of the anchor, and the banned word time point set to obtain anchor time period assessment data corresponding to each anchor in the entire live broadcast data.

[0071] In this embodiment, the entire live broadcast data and the banned word time point set are integrated according to the actual live broadcast time period of the anchor in the anchor actual live broadcast time period set to obtain the anchor time period assessment data corresponding to each anchor in the entire live broadcast data, and the anchor time period assessment data is the assessment data belonging to the corresponding anchor recorded during the entire live broadcast. Acquiring the anchor time period assessment data corresponding to the anchor based on the anchor's actual live broadcast time period can make the acquired anchor assessment data more accurate.

[0072] In one embodiment, step S16 includes:

[0073] For each anchor in the entire live broadcast data, obtaining an anchor actual live broadcast time period corresponding to the anchor in the anchor actual live broadcast time period set as a target live broadcast time period;

[0074] Acquire corresponding live broadcast period trend data and live broadcast period order data according to the target live broadcast period and the entire live broadcast data;

[0075] Acquire the number of banned words in the live broadcast period according to the target live broadcast period and the banned word time point set;

[0076] The anchor time period assessment data corresponding to the anchor is obtained according to the live broadcast time period trend data, the live broadcast time period order data and the number of banned words in the live broadcast time period.

[0077] In this embodiment, for each anchor in the entire live broadcast data, that is, each anchor working during the entire live broadcast process, first, the actual live broadcast time period corresponding to the anchor is obtained from the anchor's actual live broadcast time period as the target live broadcast time period, so as to summarize the data according to the anchor's actual live broadcast time period. Then, the entire live broadcast data includes the entire live broadcast trend data and the entire live broadcast order data, and the corresponding live broadcast time period trend data and live broadcast time period order data are obtained through the entire live broadcast trend data, the entire live broadcast order data and the target live broadcast time period, wherein the live broadcast time period trend data includes the corresponding live broadcast room entry number, the number of people leaving, the number of people joining the group, the average viewing time per person and other data in the target live broadcast time period, and the live broadcast time period order data includes the corresponding order volume, sales volume, sales volume, return rate and other data in the target live broadcast time period. In addition, according to the target live broadcast time period and the banned word time point set, the number of occurrences of banned words in the target live broadcast time period is obtained, thereby obtaining the number of banned words in the live broadcast time period. Finally, the live broadcast time period trend data, the live broadcast time period order data and the number of banned words in the live broadcast time period are summarized to obtain the anchor time period assessment data corresponding to the anchor.

[0078] S17: If an anchor performance assessment instruction is received, the anchor time period assessment data corresponding to each anchor in the anchor user information is obtained according to a preset assessment cycle, and the anchor performance assessment data set is obtained by combining.

[0079] In this embodiment, when the anchor performance appraisal instruction is received, the performance appraisal of each anchor in the anchor user information is started. Among them, the anchor performance appraisal instruction can be actively triggered when the system time reaches the target appraisal time. The target appraisal time is determined according to the preset appraisal cycle. The appraisal cycle can be set to a month, so that the anchor is appraised once a month; the anchor performance appraisal instruction can also be triggered by relevant management personnel. According to the preset appraisal cycle, the anchor time period appraisal data corresponding to each anchor in the anchor user information is obtained, and the anchor performance appraisal data set is combined to obtain the anchor performance appraisal data set, so as to quickly and accurately obtain the appraisal data required by each anchor in the anchor user information in this appraisal. Specifically, the appraisal time range is obtained according to the appraisal cycle. For example, if the appraisal cycle is a month, the appraisal time range is from the first day of the current month to the last day of the current month; for each anchor in the anchor user information, the anchor time period appraisal data corresponding to the anchor in the appraisal time range is obtained and combined into the anchor performance appraisal data of the anchor, that is, the anchor performance appraisal data is the appraisal data belonging to the anchor recorded within the appraisal time range, and the anchor performance appraisal data of each anchor in the anchor user information is combined into the anchor performance appraisal data set.

[0080] S18. Obtaining a performance appraisal result corresponding to each anchor in the anchor user information according to the performance appraisal scoring strategy and the anchor performance appraisal data set.

[0081] In this embodiment, after obtaining the anchor performance appraisal data set, the performance appraisal results corresponding to each anchor in the anchor user information can be quickly obtained through the performance appraisal scoring strategy and the anchor performance appraisal data set, so that the live broadcast process data of each anchor can be quickly and accurately automatically acquired and processed based on face recognition, voice recognition and performance appraisal scoring strategy, thereby realizing the performance appraisal of each anchor and improving the accuracy and efficiency of the anchor performance appraisal.

[0082] In one embodiment, step S18 includes:

[0083] For each anchor in the anchor user information, obtaining a point reduction result of the anchor according to the point reduction strategy and the anchor performance evaluation data set;

[0084] Obtaining the qualitative indicator result of the anchor according to the qualitative indicator scoring strategy and the anchor performance assessment data set;

[0085] Obtaining a quantitative indicator score corresponding to the anchor, and obtaining a quantitative indicator result of the anchor according to the quantitative indicator score and the quantitative indicator scoring strategy;

[0086] The performance evaluation result of the anchor is obtained according to the deduction item result, the qualitative indicator result and the quantitative indicator result.

[0087] In this embodiment, the performance appraisal scoring strategy includes a deduction strategy, a qualitative indicator scoring strategy, and a quantitative indicator scoring strategy. The deduction item results, qualitative indicator results, and quantitative indicator results of the anchor are obtained through the deduction strategy, the qualitative indicator scoring strategy, and the quantitative indicator scoring strategy, respectively, and then the final performance appraisal result is obtained according to the deduction item results, the qualitative indicator results, and the quantitative indicator results of the anchor. Preferably, the performance appraisal result can be obtained by the formula Z=V2+V3-V1, where Z is the performance appraisal result, V1 is the deduction item result, V2 is the qualitative indicator result, and V3 is the quantitative indicator result.

[0088] Specifically, for each anchor in the anchor user information, the anchor's deduction result is obtained according to the deduction strategy and the anchor performance evaluation data set, wherein the deduction strategy is generated according to the deduction information. For example, the deduction information may include the preset number of banned words and the preset deduction value corresponding to the preset number of banned words. Then the formula corresponding to the deduction strategy is V1 = (M ÷ m) × c, V1 is the deduction result, M is the total number of banned words to be obtained, m is the preset number of banned words, and c is the preset deduction value. Then, the number of banned words in all live broadcast time periods corresponding to the anchor in the anchor performance evaluation data set can be obtained, and then the total number of banned words of the anchor can be obtained. Substituting the total number of banned words into the formula corresponding to the deduction strategy can calculate the deduction result of the anchor.

[0089] The qualitative indicator results of the anchor are obtained according to the qualitative indicator scoring strategy and the anchor performance evaluation data set, wherein the qualitative indicator scoring strategy is generated according to the qualitative indicator information and the qualitative indicator weight. The qualitative indicator information may include multiple qualitative indicators, as well as the qualitative goals and scoring rules corresponding to each qualitative indicator. The qualitative indicator weight includes the qualitative indicator sub-weight corresponding to each qualitative indicator. The formula corresponding to the qualitative indicator scoring strategy is V2=k1X1+k2X2+…+k i X i , i is a positive integer and represents the serial number of the qualitative indicator, V2 is the result of the qualitative indicator, k i is the qualitative indicator sub-weight corresponding to the i-th qualitative indicator, X i is the qualitative indicator score obtained according to the qualitative target and scoring rule corresponding to the i-th qualitative indicator. For example, the first qualitative indicator is set to the sales indicator, the qualitative target corresponding to the sales indicator is the preset target amount, and the scoring rule corresponding to the sales indicator is that the single item has a full score of 100 and the score is calculated based on the achievement of the target. The scoring rule can also be understood as obtaining full marks when the actual total sales of the anchor reaches the preset target amount in the anchor performance appraisal data set. Then, according to the scoring rule, X1 = (Q ÷ q) × 100 is obtained, where Q is the actual total sales of the anchor and q is the preset target amount. For example, in one embodiment, the qualitative indicator information includes two qualitative indicators, the first qualitative indicator is the sales indicator, and the second qualitative indicator is the live broadcast duration indicator, k1 is set to 30%, and k2 is set to 20%. When the qualitative indicator score X1 corresponding to the sales indicator is 90 points according to the anchor performance assessment data set during the assessment, and the qualitative indicator score X2 corresponding to the live broadcast duration indicator is 100 points, then the qualitative indicator result V2=90×30%+100×20%=47 points.

[0090] When obtaining the quantitative indicator results of the anchor, it is necessary to first obtain the quantitative indicator score corresponding to the anchor, and then obtain the quantitative indicator results of the anchor according to the quantitative indicator score and the quantitative indicator scoring strategy. Among them, the quantitative indicator scoring strategy is based on the quantitative indicator information and the quantitative indicator weight. The quantitative indicator information may include multiple quantitative indicators, the quantitative indicator weight includes the quantitative indicator sub-weight corresponding to each quantitative indicator, the quantitative indicator score includes the quantitative indicator sub-score corresponding to each quantitative indicator, and the formula corresponding to the quantitative indicator scoring strategy is V3=b1Y1+b2Y2+…+b n Y n , n is a positive integer and represents the serial number of the quantitative indicator, V3 is the quantitative indicator result, b n is the quantitative indicator sub-weight corresponding to the nth quantitative indicator, Y nis the quantitative indicator sub-score corresponding to the nth quantitative indicator. The quantitative indicator score is the corresponding score given by the relevant managers based on the quantitative indicator information and its month-on-month information. For example, the click-through conversion rate indicator is set as the first quantitative indicator, and the click-through conversion rate of this month is 3.42%, and the click-through conversion rate of last month is 4.48%. Then, by calculating the month-on-month comparison, it is obtained that the indicator of this month has decreased by 23.66%. The relevant managers know the changes in the monthly live broadcast data based on the month-on-month situation, and give the quantitative indicator sub-score corresponding to the click-through conversion rate indicator based on the changes and experience.

[0091] The method for intelligently processing anchor user data disclosed in the present invention can automatically acquire and process the live broadcast process data of each anchor quickly and accurately, thereby improving data processing efficiency.

[0092] The embodiment of the present invention further provides a system for intelligently processing anchor user data. The system for intelligently processing anchor user data is used to execute any embodiment of the aforementioned method for intelligently processing anchor user data. Figure 2 , Figure 2 This is a schematic block diagram of an anchor user data intelligent processing system provided in one embodiment of the present invention. The anchor user data intelligent processing system provided in this embodiment of the present invention includes an information management unit 11, a performance management unit 12, a data collection unit 13, an anchor analysis unit 14, a banned word analysis unit 15, a first data integration unit 16, a second data integration unit 17, and a performance evaluation unit 18.

[0093] The information management unit 11 is used to respond to the information entry instruction and obtain the banned word information and anchor user information corresponding to the information entry instruction for storage.

[0094] In this embodiment, the information entry instruction is triggered by the relevant management personnel (such as the project manager), and the banned word information and anchor user information set by the relevant management personnel are stored through the information entry instruction. Among them, the information entry instruction includes a first information entry instruction and a second information entry instruction, the first information entry instruction is used to enter the banned word information, and the second information entry instruction is used to enter the anchor user information. There are some banned words on each live broadcast platform, such as some words suspected of deception, stimulating consumption, false propaganda, etc. The relevant management personnel set the banned words that may be risky for each live broadcast platform of the project based on their own experience, so as to input the stored banned word information according to all platform settings of the project live broadcast, so that the banned word information can be used as the reference information required for the anchor performance appraisal later. Preferably, the banned word information includes platform information, and a banned word list corresponding to the platform information, and the platform information includes the platform type and the platform live broadcast address.

[0095] Preferably, the anchor user information includes anchor portrait, anchor level (e.g., S / A / B / C level), cooperation type (e.g., full-time / part-time) and other basic information. The relevant management personnel will enter the anchor user information corresponding to the recruited anchors to facilitate the subsequent performance appraisal of the recruited anchors. In addition, the relevant management personnel can also configure the corresponding live broadcast personnel for each live broadcast platform according to the anchor user information. The live broadcast personnel include anchors and operation-related personnel, and complete the preliminary scheduling work.

[0096] The performance management unit 12 is used to respond to the performance target setting instruction and generate a corresponding performance appraisal scoring strategy according to the performance target setting instruction.

[0097] In this embodiment, the performance target setting instruction is triggered by the relevant management personnel (such as the project manager), and the relevant management personnel trigger the performance target setting instruction at the target setting time (such as the beginning of each month) to generate the corresponding performance appraisal scoring strategy.

[0098] In one embodiment, when the performance management unit 12 executes the step of generating a corresponding performance appraisal score strategy according to the performance target setting instruction, it is specifically used to:

[0099] Acquire assessment indicator information corresponding to the performance target setting instruction; wherein the assessment indicator information includes deduction item information, qualitative indicator information, qualitative indicator weight, quantitative indicator information and quantitative indicator weight;

[0100] Generate a deduction strategy according to the deduction item information;

[0101] Generating a qualitative indicator scoring strategy according to the qualitative indicator information and the qualitative indicator weight;

[0102] Generate a quantitative indicator scoring strategy according to the quantitative indicator information and the quantitative indicator weight;

[0103] The score reduction strategy, the qualitative indicator scoring strategy and the quantitative indicator scoring strategy are combined to obtain a performance appraisal scoring strategy.

[0104] In this embodiment, the performance appraisal scoring strategy includes a deduction strategy, a qualitative indicator scoring strategy, and a quantitative indicator scoring strategy. Specifically, the appraisal indicator information is used to indicate the anchor performance appraisal target set by the relevant management personnel based on the anchor user information, and the appraisal indicator information includes deduction item information, qualitative indicator information, qualitative indicator weights, quantitative indicator information, and quantitative indicator weights. Further, a deduction strategy is generated based on the deduction item information, and the deduction strategy is used to obtain the anchor's deduction item results. A qualitative indicator scoring strategy is generated based on the qualitative indicator information and the qualitative indicator weights, and the qualitative indicator scoring strategy is used to obtain the anchor's qualitative indicator results. A quantitative indicator scoring strategy is generated based on the quantitative indicator information and the quantitative indicator weights, and the quantitative indicator scoring strategy is used to obtain the anchor's quantitative indicator results.

[0105] The data collection unit 13 is used to collect the live broadcast data corresponding to the live broadcast start instruction in real time if a live broadcast start instruction is received, so as to obtain the whole live broadcast data.

[0106] In this embodiment, when the live broadcast room starts to broadcast live, the live broadcast start instruction is triggered by relevant personnel, so as to start real-time collection of the live broadcast data of the live broadcast room during the live broadcast process until the live broadcast ends, and the whole live broadcast data is obtained. Among them, the whole live broadcast data includes the whole live broadcast video data, the whole live broadcast trend data and the whole live broadcast order data. Specifically, the live broadcast video of the live broadcast room can be downloaded in real time through a preset audio and video conversion tool (such as FFMPEG tool), and the live broadcast video can be archived according to the live broadcast room and live broadcast time after the live broadcast ends, so as to obtain the whole live broadcast video data. During the live broadcast process of the live broadcast room, the number of people entering the live broadcast room, the number of people leaving, the number of people joining the group, the average viewing time per person and other data can be obtained through the system background to obtain the whole live broadcast trend data. The system background corresponds to the order management platform of each live broadcast platform, and the live broadcast order data of the live broadcast room during the live broadcast process can be obtained through SDK (Software Development Kit, Software Development Kit), such as order volume, sales volume, sales, return rate and other data, so as to obtain the whole live broadcast order data.

[0107] The anchor analysis unit 14 is used to obtain the anchor's actual live broadcast time period corresponding to each anchor in the entire live broadcast data based on the anchor user information, the entire live broadcast data and a preset face recognition model, and combine them to obtain the anchor's actual live broadcast time period set.

[0108] In this embodiment, the entire live broadcast data is used as the data source, combined with the stored anchor user information, and face recognition is performed through a face recognition model to identify the actual working time period of each anchor during the entire live broadcast, that is, the actual live broadcast time period corresponding to each anchor in the entire live broadcast data is obtained. The actual live broadcast time period represents the actual working time period of the anchor, and the actual live broadcast time period corresponding to each anchor in the entire live broadcast data is combined into an anchor actual live broadcast time period set. Since the broadcast time is a flexible working mode, the actual broadcast time will be determined within the shift according to the actual situation, so obtaining the scheduled working time period of the anchor in the scheduling work as the actual working time period may not be accurate. Therefore, the actual live broadcast time period corresponding to each anchor in the entire live broadcast data can be obtained more accurately through the anchor user information, the entire live broadcast data and the face recognition model.

[0109] In one embodiment, when the anchor analysis unit 14 executes the step of obtaining the anchor actual live broadcast time period corresponding to each anchor in the entire live broadcast data based on the anchor user information, the entire live broadcast data and the preset face recognition model, it is specifically used to:

[0110] Acquire the entire live video data from the entire live data, and acquire the entire live time period corresponding to the entire live video data;

[0111] Obtaining a detection point set according to the entire live broadcast period and a preset detection cycle;

[0112] For each detection point in the detection point set, a live broadcast picture corresponding to the detection point in the entire live broadcast video data is obtained, and an anchor identity corresponding to the detection point is obtained according to the face recognition model, the live broadcast picture and the anchor user information;

[0113] According to the anchor identity corresponding to each detection point in the detection point set, the actual live broadcast time period corresponding to each anchor in the entire live broadcast data is obtained.

[0114] In this embodiment, the entire live video data and the corresponding entire live time period in the entire live data are obtained. Face recognition is performed on the live screen of the entire live video data according to a preset detection cycle to confirm the identity of the anchor corresponding to the live screen of the detection point, and then the actual live time period corresponding to each anchor is obtained according to the anchor identity corresponding to each detection point.

[0115] Specifically, each detection point corresponding to the entire live video data can be determined through the entire live broadcast period and the preset detection cycle, and each detection point corresponding to the entire live video data can be combined to obtain a detection point set. For example, the entire live broadcast period is 10:00-18:00, and the detection cycle is 5s. Then, a detection is performed every 5s in the entire live broadcast period, and a total of 8760 detection points can be determined, and the corresponding detection point set is composed of these 8760 detection points. For each detection point in the detection point set, the live screen corresponding to the detection point in the entire live video data is obtained, and the live screen is parsed by the face recognition model and matched with the anchor avatar in the anchor user information, and then the anchor identity corresponding to the live screen of the detection point can be confirmed. After confirming the anchor identity corresponding to each detection point in the detection point set, the actual live broadcast period corresponding to each anchor in the entire live data can be obtained according to the anchor identity corresponding to each detection point.

[0116] In one embodiment, when the anchor analysis unit 14 executes the step of obtaining the anchor actual live broadcast time period corresponding to each anchor in the entire live broadcast data according to the anchor identity corresponding to each detection point in the detection point set, it is specifically used to:

[0117] A determination period set is obtained according to the entire live broadcast period and a preset determination cycle; wherein the determination period set includes a plurality of determination periods, and each determination period includes a plurality of detection points;

[0118] For each determination period in the determination period set, if it is determined that all detection points contained in the determination period correspond to the same anchor identity, then the same anchor identity corresponding to all detection points contained in the determination period is determined as the target anchor identity corresponding to the determination period;

[0119] If it is determined that the detection points contained in the determination period correspond to multiple anchor identities, the target anchor identity corresponding to the determination period is determined according to the broadcasting ratio of each anchor identity in the determination period;

[0120] The actual live broadcast time period corresponding to each anchor in the entire live broadcast data is obtained according to the target anchor identity corresponding to each determination time period in the determination time period set.

[0121] In this embodiment, after detecting the anchor identity corresponding to each detection point in the detection point set, the entire live broadcast period is divided according to the preset judgment period, so that the entire live broadcast period is divided into multiple judgment periods, and the multiple judgment periods form a judgment period set. Among them, the judgment period is greater than the detection period, so that each judgment period contains multiple detection points. For example, the detection period is 5s, the judgment period is 5 minutes, and the entire live broadcast period is divided with 5 minutes as a judgment point, then the judgment period contains 60 detection points.

[0122] Secondly, determine the anchor identity to which each determination period in the determination period set belongs. Specifically, for each determination period in the determination period set, if the detection points contained in the determination period all correspond to the same anchor identity, then the same anchor identity corresponding to the detection points contained in the determination period is determined as the target anchor identity corresponding to the determination period, that is, the determination period belongs to the same anchor identity, wherein, if the same anchor identity is an empty set, the anchor identity does not contain the anchor's identity information, indicating that no anchor user information is identified at the detection points contained in the determination period, and there is no anchor in the determination period; if the same anchor identity is a non-empty set, the anchor identity contains the anchor's identity information, indicating that the same anchor user information is identified at the detection points contained in the determination period, and the same anchor is broadcasting live in the determination period. If it is determined that the detection points contained in the judgment period correspond to multiple anchor identities, it means that there are multiple anchors broadcasting live in the judgment period, then the broadcaster appearance ratio of each anchor identity in the judgment period is obtained, and the anchor identity corresponding to the largest broadcaster appearance ratio is determined as the target anchor identity corresponding to the judgment period, that is, the judgment period is attributed to the anchor with the most appearances in this period. Finally, statistics are performed based on the target anchor identity corresponding to each judgment period in the determined judgment period set to obtain the actual broadcast period corresponding to each anchor in the entire live broadcast data.

[0123] The banned word analysis unit 15 is used to obtain the time points at which banned words appear in the entire live broadcast data based on the banned word information, the entire live broadcast data and a preset speech recognition model, and combine them to obtain a banned word time point set.

[0124] In this embodiment, the entire live broadcast data is used as the data source, combined with the stored banned word information, and speech recognition is performed through a speech recognition model to obtain the time points at which banned words appear in the entire live broadcast data, and the time points at which banned words appear in the entire live broadcast data are combined into a banned word time point set to facilitate subsequent statistics of the number of times banned words appear in each anchor's live broadcast.

[0125] In one embodiment, when the banned word analysis unit 15 executes the step of obtaining the time point at which the banned words appear in the entire live broadcast data based on the banned word information, the entire live broadcast data and a preset speech recognition model, it is specifically used to:

[0126] Using the speech recognition model to perform speech recognition on the entire live broadcast data to obtain dialogue text and text time;

[0127] The banned word information, the dialogue text and the text time are matched to obtain the time point when the banned word appears in the entire live broadcast data.

[0128] In this embodiment, the audio file of the entire live broadcast data is parsed through a speech recognition model to obtain the corresponding dialogue text and text time after conversion, and the dialogue text, text time and the stored banned word information are matched to obtain the time point when the banned word appears in the dialogue text, that is, the time point when the banned word appears in the entire live broadcast data.

[0129] The first data integration unit 16 is used to integrate the entire live broadcast data, the actual live broadcast time period set of the anchor and the banned word time point set to obtain the anchor time period assessment data corresponding to each anchor in the entire live broadcast data.

[0130] In this embodiment, the entire live broadcast data and the banned word time point set are integrated according to the actual live broadcast time period of the anchor in the anchor actual live broadcast time period set to obtain the anchor time period assessment data corresponding to each anchor in the entire live broadcast data, and the anchor time period assessment data is the assessment data belonging to the corresponding anchor recorded during the entire live broadcast. Acquiring the anchor time period assessment data corresponding to the anchor based on the anchor's actual live broadcast time period can make the acquired anchor assessment data more accurate.

[0131] In one embodiment, the first data integration unit 16 is specifically used for:

[0132] For each anchor in the entire live broadcast data, obtaining an anchor actual live broadcast time period corresponding to the anchor in the anchor actual live broadcast time period set as a target live broadcast time period;

[0133] Acquire corresponding live broadcast period trend data and live broadcast period order data according to the target live broadcast period and the entire live broadcast data;

[0134] Acquire the number of banned words in the live broadcast period according to the target live broadcast period and the banned word time point set;

[0135] The anchor time period assessment data corresponding to the anchor is obtained according to the live broadcast time period trend data, the live broadcast time period order data and the number of banned words in the live broadcast time period.

[0136] In this embodiment, for each anchor in the entire live broadcast data, that is, each anchor working during the entire live broadcast process, first, the actual live broadcast time period corresponding to the anchor is obtained from the anchor's actual live broadcast time period as the target live broadcast time period, so as to summarize the data according to the anchor's actual live broadcast time period. Then, the entire live broadcast data includes the entire live broadcast trend data and the entire live broadcast order data, and the corresponding live broadcast time period trend data and live broadcast time period order data are obtained through the entire live broadcast trend data, the entire live broadcast order data and the target live broadcast time period, wherein the live broadcast time period trend data includes the corresponding live broadcast room entry number, the number of people leaving, the number of people joining the group, the average viewing time per person and other data in the target live broadcast time period, and the live broadcast time period order data includes the corresponding order volume, sales volume, sales volume, return rate and other data in the target live broadcast time period. In addition, according to the target live broadcast time period and the banned word time point set, the number of occurrences of banned words in the target live broadcast time period is obtained, thereby obtaining the number of banned words in the live broadcast time period. Finally, the live broadcast time period trend data, the live broadcast time period order data and the number of banned words in the live broadcast time period are summarized to obtain the anchor time period assessment data corresponding to the anchor.

[0137] The second data integration unit 17 is used to obtain the anchor time period assessment data corresponding to each anchor in the anchor user information according to a preset assessment period if an anchor performance assessment instruction is received, and combine them to obtain an anchor performance assessment data set.

[0138] In this embodiment, when the anchor performance appraisal instruction is received, the performance appraisal of each anchor in the anchor user information is started. Among them, the anchor performance appraisal instruction can be actively triggered when the system time reaches the target appraisal time. The target appraisal time is determined according to the preset appraisal cycle. The appraisal cycle can be set to a month, so that the anchor is appraised once a month; the anchor performance appraisal instruction can also be triggered by relevant management personnel. According to the preset appraisal cycle, the anchor time period appraisal data corresponding to each anchor in the anchor user information is obtained, and the anchor performance appraisal data set is combined to obtain the anchor performance appraisal data set, so as to quickly and accurately obtain the appraisal data required by each anchor in the anchor user information in this appraisal. Specifically, the appraisal time range is obtained according to the appraisal cycle. For example, if the appraisal cycle is a month, the appraisal time range is from the first day of the current month to the last day of the current month; for each anchor in the anchor user information, the anchor time period appraisal data corresponding to the anchor in the appraisal time range is obtained and combined into the anchor performance appraisal data of the anchor, that is, the anchor performance appraisal data is the appraisal data belonging to the anchor recorded within the appraisal time range, and the anchor performance appraisal data of each anchor in the anchor user information is combined into the anchor performance appraisal data set.

[0139] The performance evaluation unit 18 is used to obtain the performance evaluation result corresponding to each anchor in the anchor user information according to the performance evaluation scoring strategy and the anchor performance evaluation data set.

[0140] In this embodiment, after obtaining the anchor performance appraisal data set, the performance appraisal results corresponding to each anchor in the anchor user information can be quickly obtained through the performance appraisal scoring strategy and the anchor performance appraisal data set, so that the live broadcast process data of each anchor can be quickly and accurately automatically acquired and processed based on face recognition, voice recognition and performance appraisal scoring strategy, thereby realizing the performance appraisal of each anchor and improving the accuracy and efficiency of the anchor performance appraisal.

[0141] In one embodiment, the performance evaluation unit 18 is specifically used for:

[0142] For each anchor in the anchor user information, obtaining a point reduction result of the anchor according to the point reduction strategy and the anchor performance evaluation data set;

[0143] Obtaining the qualitative indicator result of the anchor according to the qualitative indicator scoring strategy and the anchor performance assessment data set;

[0144] Obtaining a quantitative indicator score corresponding to the anchor, and obtaining a quantitative indicator result of the anchor according to the quantitative indicator score and the quantitative indicator scoring strategy;

[0145] The performance evaluation result of the anchor is obtained according to the deduction item result, the qualitative indicator result and the quantitative indicator result.

[0146] In this embodiment, the performance appraisal scoring strategy includes a deduction strategy, a qualitative indicator scoring strategy, and a quantitative indicator scoring strategy. The deduction item results, qualitative indicator results, and quantitative indicator results of the anchor are obtained through the deduction strategy, the qualitative indicator scoring strategy, and the quantitative indicator scoring strategy, respectively, and then the final performance appraisal result is obtained according to the deduction item results, the qualitative indicator results, and the quantitative indicator results of the anchor. Preferably, the performance appraisal result can be obtained by the formula Z=V2+V3-V1, where Z is the performance appraisal result, V1 is the deduction item result, V2 is the qualitative indicator result, and V3 is the quantitative indicator result.

[0147] Specifically, for each anchor in the anchor user information, the anchor's deduction result is obtained according to the deduction strategy and the anchor performance evaluation data set, wherein the deduction strategy is generated according to the deduction information. For example, the deduction information may include the preset number of banned words and the preset deduction value corresponding to the preset number of banned words. Then the formula corresponding to the deduction strategy is V1 = (M ÷ m) × c, V1 is the deduction result, M is the total number of banned words to be obtained, m is the preset number of banned words, and c is the preset deduction value. Then, the number of banned words in all live broadcast time periods corresponding to the anchor in the anchor performance evaluation data set can be obtained, and then the total number of banned words of the anchor can be obtained. Substituting the total number of banned words into the formula corresponding to the deduction strategy can calculate the deduction result of the anchor.

[0148] The qualitative indicator results of the anchor are obtained according to the qualitative indicator scoring strategy and the anchor performance evaluation data set, wherein the qualitative indicator scoring strategy is generated according to the qualitative indicator information and the qualitative indicator weight. The qualitative indicator information may include multiple qualitative indicators, as well as the qualitative goals and scoring rules corresponding to each qualitative indicator. The qualitative indicator weight includes the qualitative indicator sub-weight corresponding to each qualitative indicator. The formula corresponding to the qualitative indicator scoring strategy is V2=k1X1+k2X2+…+k i X i , i is a positive integer and represents the serial number of the qualitative indicator, V2 is the result of the qualitative indicator, k i is the qualitative indicator sub-weight corresponding to the i-th qualitative indicator, X i is the qualitative indicator score obtained according to the qualitative target and scoring rule corresponding to the i-th qualitative indicator. For example, the first qualitative indicator is set to the sales indicator, the qualitative target corresponding to the sales indicator is the preset target amount, and the scoring rule corresponding to the sales indicator is that the single item has a full score of 100 and the score is calculated based on the achievement of the target. The scoring rule can also be understood as obtaining full marks when the actual total sales of the anchor reaches the preset target amount in the anchor performance appraisal data set. Then, according to the scoring rule, X1 = (Q ÷ q) × 100 is obtained, where Q is the actual total sales of the anchor and q is the preset target amount. For example, in one embodiment, the qualitative indicator information includes two qualitative indicators, the first qualitative indicator is the sales indicator, and the second qualitative indicator is the live broadcast duration indicator, k1 is set to 30%, and k2 is set to 20%. When the qualitative indicator score X1 corresponding to the sales indicator is 90 points according to the anchor performance assessment data set during the assessment, and the qualitative indicator score X2 corresponding to the live broadcast duration indicator is 100 points, then the qualitative indicator result V2=90×30%+100×20%=47 points.

[0149] When obtaining the quantitative indicator results of the anchor, it is necessary to first obtain the quantitative indicator score corresponding to the anchor, and then obtain the quantitative indicator results of the anchor according to the quantitative indicator score and the quantitative indicator scoring strategy. Among them, the quantitative indicator scoring strategy is based on the quantitative indicator information and the quantitative indicator weight. The quantitative indicator information may include multiple quantitative indicators, the quantitative indicator weight includes the quantitative indicator sub-weight corresponding to each quantitative indicator, the quantitative indicator score includes the quantitative indicator sub-score corresponding to each quantitative indicator, and the formula corresponding to the quantitative indicator scoring strategy is V3=b1Y1+b2Y2+…+b n Y n , n is a positive integer and represents the serial number of the quantitative indicator, V3 is the quantitative indicator result, b n is the quantitative indicator sub-weight corresponding to the nth quantitative indicator, Y n is the quantitative indicator sub-score corresponding to the nth quantitative indicator. The quantitative indicator score is the corresponding score given by the relevant managers based on the quantitative indicator information and its month-on-month information. For example, the click-through conversion rate indicator is set as the first quantitative indicator, and the click-through conversion rate of this month is 3.42%, and the click-through conversion rate of last month is 4.48%. Then, by calculating the month-on-month comparison, it is obtained that the indicator of this month has decreased by 23.66%. The relevant managers know the changes in the monthly live broadcast data based on the month-on-month situation, and give the quantitative indicator sub-score corresponding to the click-through conversion rate indicator based on the changes and experience.

[0150] The present invention discloses an intelligent system for processing anchor user data, which is used to execute any embodiment of the aforementioned intelligent method for processing anchor user data, and can automatically acquire and process the live broadcast process data of each anchor quickly and accurately, thereby improving data processing efficiency.

[0151] The above-mentioned method for intelligently processing anchor user data can be implemented in the form of a computer program. The computer program can be used in Figure 3 Runs on the computer device shown.

[0152] See also Figure 3 , Figure 3 5 is a schematic block diagram of a computer device provided by an embodiment of the present invention. The computer device 500 includes a processor 502 , a memory and a network interface 505 connected via a device bus 501 , wherein the memory may include a storage medium 503 and an internal memory 504 .

[0153] The storage medium 503 can store an operating device 5031 and a computer program 5032. When the computer program 5032 is executed, the processor 502 can execute the anchor user data intelligent processing method.

[0154] The processor 502 is used to provide computing and control capabilities to support the operation of the entire computer device 500 .

[0155] The internal memory 504 provides an environment for the operation of the computer program 5032 in the storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute the intelligent processing method for the anchor user data.

[0156] The network interface 505 is used for network communication, such as providing data information transmission, etc. Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device 500 to which the solution of the present invention is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0157] The processor 502 is used to run the computer program 5032 stored in the memory to implement the method for intelligently processing anchor user data disclosed in the embodiment of the present invention.

[0158] Those skilled in the art will understand that Figure 3 The embodiments of the computer device shown in the figure do not constitute a limitation on the specific composition of the computer device. In other embodiments, the computer device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. For example, in some embodiments, the computer device may only include a memory and a processor. In such embodiments, the structure and function of the memory and the processor are the same as those of the embodiment of the present invention. Figure 3 The embodiments shown are consistent and will not be described again here.

[0159] It should be understood that in the embodiment of the present invention, the processor 502 may be a central processing unit (CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0160] In another embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the method for intelligently processing anchor user data disclosed in the embodiment of the present invention is implemented.

[0161] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the above-described equipment and units can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here. Those of ordinary skill in the art can appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are executed 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 exceed the scope of the present invention.

[0162] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. Units with the same function may also be combined into one unit. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, systems or units, or may be electrical, mechanical or other forms of connection.

[0163] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention.

[0164] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0165] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a background server, or a network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), disk or optical disk and other media that can store program codes.

[0166] The above are only specific embodiments 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 various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for intelligently processing anchor user data, characterized in that: include: In response to an information entry instruction, obtaining banned word information and anchor user information corresponding to the information entry instruction for storage; In response to a performance target setting instruction, generating a corresponding performance appraisal scoring strategy according to the performance target setting instruction; If a live broadcast start instruction is received, live broadcast data corresponding to the live broadcast start instruction is collected in real time to obtain the whole live broadcast data; Based on the anchor user information, the entire live broadcast data and a preset face recognition model, the actual live broadcast time period corresponding to each anchor in the entire live broadcast data is obtained, and the actual live broadcast time period set of the anchor is obtained by combining; Based on the banned word information, the entire live broadcast data and a preset speech recognition model, the time points at which banned words appear in the entire live broadcast data are obtained, and a banned word time point set is obtained by combining them; The anchor time period assessment data corresponding to each anchor in the entire live broadcast data is obtained by integrating the entire live broadcast data, the anchor's actual live broadcast time period set and the banned word time point set; If an anchor performance assessment instruction is received, the anchor time period assessment data corresponding to each anchor in the anchor user information is obtained according to a preset assessment period, and the anchor performance assessment data set is obtained by combining them; The performance appraisal result corresponding to each anchor in the anchor user information is obtained according to the performance appraisal scoring strategy and the anchor performance appraisal data set.

2. The method for intelligently processing anchor user data according to claim 1, characterized in that: The acquiring the actual live broadcast time period corresponding to each anchor in the entire live broadcast data based on the anchor user information, the entire live broadcast data and a preset face recognition model includes: Obtaining the entire live video data from the entire live data, and obtaining the entire live time period corresponding to the entire live video data; Obtaining a detection point set according to the entire live broadcast period and a preset detection cycle; For each detection point in the detection point set, a live broadcast picture corresponding to the detection point in the entire live broadcast video data is obtained, and an anchor identity corresponding to the detection point is obtained according to the face recognition model, the live broadcast picture and the anchor user information; According to the anchor identity corresponding to each detection point in the detection point set, the actual live broadcast time period corresponding to each anchor in the entire live broadcast data is obtained.

3. The method for intelligently processing anchor user data according to claim 2, characterized in that: The obtaining, according to the anchor identity corresponding to each detection point in the detection point set, the actual live broadcast time period corresponding to each anchor in the entire live broadcast data, includes: A determination period set is obtained according to the entire live broadcast period and a preset determination cycle; wherein the determination period set includes a plurality of determination periods, and each determination period includes a plurality of detection points; For each determination period in the determination period set, if it is determined that all detection points contained in the determination period correspond to the same anchor identity, then the same anchor identity corresponding to all detection points contained in the determination period is determined as the target anchor identity corresponding to the determination period; If it is determined that the detection points contained in the determination period correspond to multiple anchor identities, the target anchor identity corresponding to the determination period is determined according to the broadcasting ratio of each anchor identity in the determination period; The actual live broadcast time period corresponding to each anchor in the entire live broadcast data is obtained according to the target anchor identity corresponding to each determination time period in the determination time period set.

4. The method for intelligently processing anchor user data according to claim 1, characterized in that: The step of obtaining the time point at which the banned word appears in the entire live broadcast data based on the banned word information, the entire live broadcast data, and a preset speech recognition model includes: Using the speech recognition model to perform speech recognition on the entire live broadcast data to obtain dialogue text and text time; The banned word information, the dialogue text and the text time are matched to obtain the time point when the banned word appears in the entire live broadcast data.

5. The method for intelligently processing anchor user data according to claim 1, characterized in that: The anchor time period assessment data corresponding to each anchor in the entire live broadcast data is obtained by integrating the entire live broadcast data, the anchor's actual live broadcast time period set and the banned word time point set, including: For each anchor in the entire live broadcast data, obtaining an anchor actual live broadcast time period corresponding to the anchor in the anchor actual live broadcast time period set as a target live broadcast time period; Acquire corresponding live broadcast period trend data and live broadcast period order data according to the target live broadcast period and the entire live broadcast data; Acquire the number of banned words in the live broadcast period according to the target live broadcast period and the banned word time point set; The anchor time period assessment data corresponding to the anchor is obtained according to the live broadcast time period trend data, the live broadcast time period order data and the number of banned words in the live broadcast time period.

6. The method for intelligently processing anchor user data according to claim 1, characterized in that: Generating a corresponding performance appraisal scoring strategy according to the performance target setting instruction includes: Acquire assessment indicator information corresponding to the performance target setting instruction; wherein the assessment indicator information includes deduction item information, qualitative indicator information, qualitative indicator weight, quantitative indicator information and quantitative indicator weight; Generate a deduction strategy according to the deduction item information; Generating a qualitative indicator scoring strategy according to the qualitative indicator information and the qualitative indicator weight; Generate a quantitative indicator scoring strategy according to the quantitative indicator information and the quantitative indicator weight; The score reduction strategy, the qualitative indicator scoring strategy and the quantitative indicator scoring strategy are combined to obtain a performance appraisal scoring strategy.

7. The method for intelligently processing anchor user data according to claim 6, characterized in that: The obtaining, according to the performance assessment scoring strategy and the anchor performance assessment data set, a performance assessment result corresponding to each anchor in the anchor user information includes: For each anchor in the anchor user information, obtaining a point reduction result of the anchor according to the point reduction strategy and the anchor performance evaluation data set; Obtaining the qualitative indicator result of the anchor according to the qualitative indicator scoring strategy and the anchor performance assessment data set; Obtaining a quantitative indicator score corresponding to the anchor, and obtaining a quantitative indicator result of the anchor according to the quantitative indicator score and the quantitative indicator scoring strategy; The performance evaluation result of the anchor is obtained according to the deduction item result, the qualitative indicator result and the quantitative indicator result.

8. An intelligent processing system for anchor user data, characterized in that: include: An information management unit, configured to, in response to an information entry instruction, obtain and store banned word information and anchor user information corresponding to the information entry instruction; A performance management unit, configured to respond to a performance target setting instruction and generate a corresponding performance appraisal scoring strategy according to the performance target setting instruction; A data collection unit, for collecting live broadcast data corresponding to the live broadcast start instruction in real time if a live broadcast start instruction is received, so as to obtain the whole live broadcast data; The anchor analysis unit is used to obtain the anchor's actual live broadcast time period corresponding to each anchor in the entire live broadcast data based on the anchor user information, the entire live broadcast data and a preset face recognition model, and combine them to obtain an anchor's actual live broadcast time period set; A banned word analysis unit, configured to obtain the time points at which banned words appear in the entire live broadcast data based on the banned word information, the entire live broadcast data, and a preset speech recognition model, and combine them to obtain a banned word time point set; A first data integration unit is used to integrate the entire live broadcast data, the actual live broadcast time period set of the anchor and the banned word time point set to obtain the anchor time period assessment data corresponding to each anchor in the entire live broadcast data; The second data integration unit is used for obtaining the anchor time period assessment data corresponding to each anchor in the anchor user information according to a preset assessment period if an anchor performance assessment instruction is received, and combining them to obtain an anchor performance assessment data set; The performance evaluation unit is used to obtain the performance evaluation result corresponding to each anchor in the anchor user information according to the performance evaluation scoring strategy and the anchor performance evaluation data set.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for intelligently processing anchor user data according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, enables the processor to execute the method for intelligent processing of anchor user data as described in any one of claims 1 to 7.

Images