A cloud platform customer management system and method
Through the cloud platform customer management system, the evaluation standards of orders and usage records are used to provide customers with personalized recommendations of paid functions, solving the accuracy problem of customers when choosing paid functions and improving experience and satisfaction.
Patent Information
- Application Number
- CN202111259755.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-10-28
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2041-10-28
AI Technical Summary
Cloud platform customers lack accuracy and humanization when choosing paid functions, resulting in customers being unable to effectively choose appropriate functions, affecting the experience.
Design a cloud platform customer management system, obtain customer orders and usage records, evaluate based on preset evaluation standards, and generate recommended content to provide customers with accurate paid function suggestions. The system includes identification modules, evaluation modules and recommendation modules, leveraging order types, usage durations and evaluation standards libraries to evaluate customer needs and provide personalized recommendations.
Improve customer experience, help customers accurately choose paid functions on the cloud platform, and improve customer satisfaction.
Smart Images

Figure CN114140152B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of cloud platforms, and in particular to a cloud platform customer management system and method. Background Art
[0002] At present, cloud platform customers will not choose paid functions on the cloud platform (such as cloud server CPU, image, memory, etc.) at the beginning. They will choose a shorter subscription period, try it out first, and then choose the appropriate functions. However, it is not accurate and humane for customers to choose the appropriate functions by themselves.
[0003] Therefore, a solution is urgently needed. Summary of the invention
[0004] One of the purposes of the present invention is to provide a cloud platform customer management system and method, which can evaluate the first usage record of the first customer, and based on the evaluation results, accurately recommend to the customer how to choose the paid functions on the cloud platform, thereby improving the customer experience and being very user-friendly.
[0005] An embodiment of the present invention provides a cloud platform customer management system, including:
[0006] A determination module, configured to obtain order information of a first customer and determine an evaluation criterion based on the order information;
[0007] an evaluation module, configured to obtain a first usage record of the first customer when the duration of use of the cloud platform by the first customer reaches a preset duration threshold, and evaluate the first usage record based on the evaluation criteria;
[0008] The recommendation module is used to generate recommendation content based on the evaluation result and recommend it to the first customer.
[0009] Preferably, the determining module performs the following operations:
[0010] Extracting multiple order items from the order information, the order items including: order type and order duration;
[0011] Based on a preset subscription type-subscription duration-evaluation standard library, determining an evaluation standard corresponding to the subscription type and the subscription duration;
[0012] Integrate the above-mentioned evaluation criteria and complete the determination.
[0013] Preferably, the evaluation module performs the following operations:
[0014] extracting a target record corresponding to the subscription type from the first usage record;
[0015] Based on the evaluation criteria, the corresponding target records are evaluated to obtain evaluation results;
[0016] Integrate the evaluation results to complete the evaluation.
[0017] Preferably, the recommendation module performs the following operations:
[0018] Extracting multiple evaluation items from the evaluation results, wherein the evaluation items include: evaluation type and score;
[0019] Determine the recommended content corresponding to the evaluation type and the score based on a preset evaluation type-score-recommended content library;
[0020] The recommended contents are integrated to complete the generation, and at the same time, are recommended to the first customer.
[0021] Preferably, the cloud platform customer management system further includes:
[0022] A compensation module, used for compensating the first customer accordingly based on the evaluation result;
[0023] The compensation module performs the following operations:
[0024] Based on a preset evaluation type-score-compensation method library, try to determine a compensation method corresponding to the evaluation type and the score;
[0025] Based on the compensation method, the first customer is compensated.
[0026] Preferably, the cloud platform customer management system further includes:
[0027] a construction module, configured to construct an association area when the first customer accesses the function module of the cloud platform, and associate an appropriate second customer with the first customer in the association area, so as to facilitate communication between the first customer and the second customer in the association area;
[0028] The building block performs the following operations:
[0029] Acquire a preset online customer set, the online customer set comprising: a plurality of third customers;
[0030] confirming whether the third client is accessing the functional module;
[0031] If so, the corresponding third customer is regarded as the fourth customer;
[0032] respectively obtaining first attribute information of the first customer and second attribute information of the fourth customer;
[0033] Based on a preset function module-key attribute library, determining at least one key attribute corresponding to the function module;
[0034] Determining a plurality of first attribute items corresponding to the key attribute in the first attribute information;
[0035] Determining a plurality of second attribute items corresponding to the key attribute in the second attribute information;
[0036] Performing feature extraction on the first attribute item to obtain a plurality of first features;
[0037] Performing feature extraction on the second attribute item to obtain a plurality of second features;
[0038] Matching the first feature with the second feature to obtain a first matching degree;
[0039] Summarizing the first matching degrees to obtain a first matching degree sum, and matching it with the corresponding key type;
[0040] Based on a preset key attribute-matching degree-association determination value library, determining a first association determination value corresponding to the key attribute and the first matching degree;
[0041] Summarizing the first association determination values to obtain a sum of association determination values;
[0042] If the sum of the association determination values is greater than or equal to a preset first threshold, the fourth customer is regarded as the fifth customer;
[0043] Obtaining a second usage record of the fifth customer;
[0044] Based on a preset function module-key record type library, determining at least one key record type corresponding to the function module;
[0045] determining a plurality of first record items corresponding to the key record type in the first usage record;
[0046] determining a plurality of second record items corresponding to the key record type in the second usage record;
[0047] Establish a first time axis and a second time axis respectively;
[0048] Based on a preset first setting rule, the first record item is set on the first time axis, and at the same time, the second record item is set on the second time axis;
[0049] Randomly extracting features from one of the first record items to obtain a plurality of third features;
[0050] Determining a first position of the first record item for feature extraction on the first time axis;
[0051] Determine the second record item within a preset first range before and / or after a second position corresponding to the first position on the second time axis, and use it as the third record item;
[0052] Performing feature extraction on the third record item to obtain a plurality of fourth features;
[0053] Matching the third feature with the fourth feature to obtain a second matching degree;
[0054] Summarizing the second matching degrees to obtain a second matching degree sum, and matching it with the corresponding key record type;
[0055] Based on the preset key record type-matching degree and-value degree, determining the key record type and the second matching degree and the first value degree corresponding to each other;
[0056] If the first value is greater than or equal to a preset second threshold, determining a third position corresponding to the third record item on the second time axis;
[0057] Setting a selection direction, the selection direction includes: front or back;
[0058] Selecting the first record item in the selection direction of the first record item for feature extraction on the first time axis as the fourth record item;
[0059] Determining a fourth position of the fourth record item on the first time axis;
[0060] Determine the second record item within a preset second range before and / or after a fifth position corresponding to the fourth position on the second time axis, and use the second record item as the fifth record item;
[0061] Performing feature extraction on the fourth record item to obtain a plurality of fifth features;
[0062] Performing feature extraction on the fifth record item to obtain a plurality of sixth features;
[0063] Matching the fifth feature with the sixth feature to obtain a third matching degree;
[0064] Summarizing the third matching degrees to obtain a third matching degree sum;
[0065] Based on a preset key record type-selection direction-matching degree and-value degree library, determine the key record type, selection direction and third matching degree and the corresponding second value degree;
[0066] Summarize the first value and the second value to obtain a second association determination value, and correspond it to the corresponding key record type;
[0067] Summarizing the second association determination values to obtain a second association determination value sum;
[0068] If the sum of the second association determination value is greater than or equal to a preset third threshold, the corresponding fifth customer will be regarded as the second customer.
[0069] Preferably, the cloud platform customer management system further includes:
[0070] A monitoring module, used to monitor whether the first customer complies with the regulations when accessing the cloud platform, and if not, execute corresponding measures;
[0071] The monitoring module performs the following operations:
[0072] Periodically acquiring a plurality of first behavior items newly generated by the first customer;
[0073] Establishing a third timeline;
[0074] Based on a preset second setting rule, setting the first behavior item on the third time axis;
[0075] Extracting features from the first behavior items one by one from the starting point to the end point of the third time axis to obtain a plurality of seventh features;
[0076] Obtaining a preset trigger feature pair library, and randomly selecting a trigger feature pair from the trigger feature pair library, wherein the trigger feature pair includes: a first trigger feature and at least one second trigger feature that correspond to each other;
[0077] matching the seventh feature with the first trigger feature;
[0078] If the match is satisfied, determining the number of the second trigger features corresponding to the first trigger features that match the match;
[0079] Selecting the number of the first behavior items after the first behavior item for feature extraction on the third time axis as the second behavior items;
[0080] Performing feature extraction on the second behavior item to obtain a plurality of eighth features;
[0081] Matching the eighth feature with the second trigger feature corresponding to the first trigger feature that matches;
[0082] If the match is met, the second trigger feature that matches the match is used as the target feature, and at the same time, the sixth position of the second behavior item corresponding to the eighth feature that matches the match on the third time axis is determined;
[0083] Query the preset trigger feature-acquisition direction-selection range-malicious behavior library to determine the acquisition direction, selection range and malicious behavior corresponding to the target feature;
[0084] Selecting the first behavior item within the selected range in the acquisition direction of the sixth position on the third time axis as the third behavior item;
[0085] If the third behavior item includes the malicious behavior, determining that the first client's access is unqualified;
[0086] Determine the measures corresponding to the malicious behavior based on a preset malicious behavior-measure library;
[0087] Carry out the measures described.
[0088] An embodiment of the present invention provides a cloud platform customer management method, comprising:
[0089] Step S1: Obtain order information of a first customer, and determine evaluation criteria based on the order information;
[0090] Step S2: when the duration of the first customer's use of the cloud platform reaches a preset duration threshold, obtaining a first usage record of the first customer, and evaluating the first usage record based on the evaluation criteria;
[0091] Step S3: Generate recommended content based on the evaluation results and recommend it to the first customer.
[0092] Preferably, in step S1, determining the evaluation criteria based on the order information includes:
[0093] Extracting multiple order items from the order information, the order items including: order type and order duration;
[0094] Based on a preset subscription type-subscription duration-evaluation standard library, determining an evaluation standard corresponding to the subscription type and the subscription duration;
[0095] Integrate the above-mentioned evaluation criteria and complete the determination.
[0096] Preferably, evaluating the first usage record based on the evaluation criteria includes:
[0097] extracting a target record corresponding to the subscription type from the first usage record;
[0098] Based on the evaluation criteria, the corresponding target records are evaluated to obtain evaluation results;
[0099] Integrate the evaluation results to complete the evaluation.
[0100] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.
[0101] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0102] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:
[0103] Figure 1 A schematic diagram of a cloud platform customer management system according to an embodiment of the present invention;
[0104] Figure 2 This is a flow chart of a cloud platform customer management method in an embodiment of the present invention;
[0105] Figure 3 The present invention is a flowchart of another cloud platform customer management method in an embodiment of the present invention. DETAILED DESCRIPTION
[0106] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.
[0107] The embodiment of the present invention provides a cloud platform customer management system, such as Figure 1 As shown, including:
[0108] Determination module 1, used to obtain order information of a first customer, and determine evaluation criteria based on the order information;
[0109] Evaluation module 2, configured to obtain a first usage record of the first customer when the duration of use of the cloud platform by the first customer reaches a preset duration threshold, and evaluate the first usage record based on the evaluation criteria;
[0110] The recommendation module 3 is used to generate recommendation content based on the evaluation results and recommend it to the first customer.
[0111] The working principle and beneficial effects of the above technical solution are:
[0112] Obtain the order information of the first customer (for example, an 8-core CPU is ordered for a period of 1 month); determine the evaluation criteria based on the order information (for example, whether the computing power of the 8-core CPU can meet the usage requirements of the first customer); when the usage time of the first customer using the cloud platform reaches a preset time threshold (for example, 20 days), obtain the usage record of the first customer, and evaluate the usage record based on the evaluation criteria; generate recommendation content (recommend upgrading the CPU to 12 cores) based on the evaluation result (for example, the computing power of the 8-core CPU cannot meet the usage requirements of the first customer);
[0113] The embodiment of the present invention can evaluate the first usage record of the first customer, and based on the evaluation result, accurately recommend to the customer how to choose the paid function on the cloud platform, thereby improving the customer experience and being very user-friendly.
[0114] The embodiment of the present invention provides a cloud platform customer management system, wherein the determination module 1 performs the following operations:
[0115] Extracting multiple order items from the order information, the order items including: order type and order duration;
[0116] Based on a preset subscription type-subscription duration-evaluation standard library, determining an evaluation standard corresponding to the subscription type and the subscription duration;
[0117] Integrate the above-mentioned evaluation criteria and complete the determination.
[0118] The working principle and beneficial effects of the above technical solution are:
[0119] The preset order type-order duration-evaluation standard library is specifically: a database containing evaluation standards corresponding to different order types and different order durations. For example, if the order type is to order an 8-core CPU and the order duration is 15 days, the corresponding evaluation standard is whether the average level of the 8-core CPU within 15 days meets the customer's business processing requirements;
[0120] Based on the order type, order duration and evaluation standard library, determine the evaluation standards, and complete the determination by integration.
[0121] The embodiment of the present invention provides a cloud platform customer management system, wherein the evaluation module 2 performs the following operations:
[0122] extracting a target record corresponding to the subscription type from the first usage record;
[0123] Based on the evaluation criteria, the corresponding target records are evaluated to obtain evaluation results;
[0124] Integrate the evaluation results to complete the evaluation.
[0125] The working principle and beneficial effects of the above technical solution are:
[0126] For example, if the order type is to order an 8-core CPU, the target record corresponding to the order type in the first usage record is extracted, that is, the load condition (for example, overload) of the 8-core CPU when the customer performs business processing;
[0127] Based on the evaluation criteria, the target records are evaluated to complete the evaluation.
[0128] The embodiment of the present invention provides a cloud platform customer management system, wherein the recommendation module 3 performs the following operations:
[0129] Extracting multiple evaluation items from the evaluation results, wherein the evaluation items include: evaluation type and score;
[0130] Determine the recommended content corresponding to the evaluation type and the score based on a preset evaluation type-score-recommended content library;
[0131] The recommended contents are integrated to complete the generation, and at the same time, are recommended to the first customer.
[0132] The working principle and beneficial effects of the above technical solution are:
[0133] The preset evaluation type-score-recommendation content library is specifically: including recommended content corresponding to different evaluation types and different scores. For example: the evaluation type is that the CPU cannot meet the requirements, and the score is 10 (the lower the score, the more serious the situation where the CPU cannot meet the requirements), and the corresponding recommended content is that it is recommended to upgrade the CPU to 12 cores.
[0134] An embodiment of the present invention provides a cloud platform customer management system, further comprising:
[0135] A compensation module, used for compensating the first customer accordingly based on the evaluation result;
[0136] The compensation module performs the following operations:
[0137] Based on a preset evaluation type-score-compensation method library, try to determine a compensation method corresponding to the evaluation type and the score;
[0138] Based on the compensation method, the first customer is compensated.
[0139] The working principle and beneficial effects of the above technical solution are:
[0140] The preset evaluation type-score-compensation method library specifically includes compensation methods corresponding to different evaluation types and different scores. For example, if the evaluation type is that the CPU cannot meet the demand and the score is 10, the corresponding compensation method is a free trial of a 12-core CPU.
[0141] An embodiment of the present invention provides a cloud platform customer management system, further comprising:
[0142] a construction module, configured to construct an association area when the first customer accesses the function module of the cloud platform, and associate an appropriate second customer with the first customer in the association area, so as to facilitate communication between the first customer and the second customer in the association area;
[0143] The building block performs the following operations:
[0144] Acquire a preset online customer set, the online customer set comprising: a plurality of third customers;
[0145] confirming whether the third client is accessing the functional module;
[0146] If so, the corresponding third customer is regarded as the fourth customer;
[0147] respectively obtaining first attribute information of the first customer and second attribute information of the fourth customer;
[0148] Based on a preset function module-key attribute library, determining at least one key attribute corresponding to the function module;
[0149] Determining a plurality of first attribute items corresponding to the key attribute in the first attribute information;
[0150] Determining a plurality of second attribute items corresponding to the key attribute in the second attribute information;
[0151] Performing feature extraction on the first attribute item to obtain a plurality of first features;
[0152] Performing feature extraction on the second attribute item to obtain a plurality of second features;
[0153] Matching the first feature with the second feature to obtain a first matching degree;
[0154] Summarizing the first matching degrees to obtain a first matching degree sum, and matching it with the corresponding key type;
[0155] Based on a preset key attribute-matching degree-association determination value library, determining a first association determination value corresponding to the key attribute and the first matching degree;
[0156] Summarizing the first association determination values to obtain a sum of association determination values;
[0157] If the sum of the association determination values is greater than or equal to a preset first threshold, the fourth customer is regarded as the fifth customer;
[0158] Obtaining a second usage record of the fifth customer;
[0159] Based on a preset function module-key record type library, determining at least one key record type corresponding to the function module;
[0160] determining a plurality of first record items corresponding to the key record type in the first usage record;
[0161] determining a plurality of second record items corresponding to the key record type in the second usage record;
[0162] Establish a first time axis and a second time axis respectively;
[0163] Based on a preset first setting rule, the first record item is set on the first time axis, and at the same time, the second record item is set on the second time axis;
[0164] Randomly extracting features from one of the first record items to obtain a plurality of third features;
[0165] Determining a first position of the first record item for feature extraction on the first time axis;
[0166] Determine the second record item within a preset first range before and / or after a second position corresponding to the first position on the second time axis, and use it as the third record item;
[0167] Performing feature extraction on the third record item to obtain a plurality of fourth features;
[0168] Matching the third feature with the fourth feature to obtain a second matching degree;
[0169] Summarizing the second matching degrees to obtain a second matching degree sum, and matching it with the corresponding key record type;
[0170] Based on the preset key record type-matching degree and-value degree, determining the key record type and the second matching degree and the first value degree corresponding to each other;
[0171] If the first value is greater than or equal to a preset second threshold, determining a third position corresponding to the third record item on the second time axis;
[0172] Setting a selection direction, the selection direction includes: front or back;
[0173] Selecting the first record item in the selection direction of the first record item for feature extraction on the first time axis as the fourth record item;
[0174] Determining a fourth position of the fourth record item on the first time axis;
[0175] Determine the second record item within a preset second range before and / or after a fifth position corresponding to the fourth position on the second time axis, and use the second record item as the fifth record item;
[0176] Performing feature extraction on the fourth record item to obtain a plurality of fifth features;
[0177] Performing feature extraction on the fifth record item to obtain a plurality of sixth features;
[0178] Matching the fifth feature with the sixth feature to obtain a third matching degree;
[0179] Summarizing the third matching degrees to obtain a third matching degree sum;
[0180] Based on a preset key record type-selection direction-matching degree and-value degree library, determine the key record type, selection direction and third matching degree and the corresponding second value degree;
[0181] Summarize the first value and the second value to obtain a second association determination value, and correspond it to the corresponding key record type;
[0182] Summarizing the second association determination values to obtain a second association determination value sum;
[0183] If the sum of the second association determination value is greater than or equal to a preset third threshold, the corresponding fifth customer will be regarded as the second customer.
[0184] The working principle and beneficial effects of the above technical solution are:
[0185] The preset online customer set is specifically: a collection of multiple online customers (third customers); the preset function module-key attribute library is specifically: a database containing key attributes corresponding to different function modules, for example: the function module is a storage function area (providing cloud storage function), the key attribute is the customer's usage scenario (used to store department information, the number of department staff is 10), etc.; the preset key attribute-matching degree and-association judgment value library is specifically: containing different key attributes and different matching degrees and corresponding association judgment values, the larger the matching degree and the larger the association judgment value, for example: the key attribute is the customer's usage scenario, the first matching degree and the corresponding association judgment value is 87; the preset first threshold is specifically: for example, 820; the preset function module-key record type library is specifically: a database containing key record types corresponding to different function modules, for example: the function module is a storage function area, and the key record type is daily upload volume (the amount of data uploaded by the customer to the cloud storage area every day); the preset first setting rule is specifically: the generation time of the record item and the time axis are The time nodes are set one by one; the preset first range is specifically: the range value on the time axis, for example: 0.5 days; the preset key record type-matching degree and-value are specifically: a database containing different key record types, different matching degrees and corresponding first values. The greater the matching degree, the greater the value. For example: the key record type is daily upload volume, the second matching degree is 50, and the first value is 66; the preset second threshold is specifically: for example, 750; the preset second range is specifically: the range value on the time axis, for example: 0.6 days; the preset key record type-selection direction-matching degree and-value database is specifically: containing different key record types, different selection directions and different third matching degrees and corresponding values. Different selection directions also affect the value. For example: the usage record is daily upload volume. The more the daily upload volume of the first customer and the second customer (the selection direction is later) matches, the more their usage is similar, the higher the value, and the previous daily upload volume has a lower value; the preset third threshold is specifically: for example, 800;
[0186] When the cloud platform is actually used, even if it recommends to customers how to choose paid functions, since customers all hope to save costs, how to use the functions more reasonably is more important. Due to the influence of customers, the platform cannot obtain customer data, but customers can communicate with each other. For example, the first customer and the second customer are both financial directors of companies of similar size, and store invoices, payrolls, etc. on the cloud platform. However, how to compress the invoices to ensure that the size of the data is minimized before uploading, the first customer and the second customer communicate to better and reasonably use the storage function of the cloud platform; therefore, this application constructs an association area (a functional submodule that can be displayed as a floating window when the customer accesses the functional module) for the first customer The second customer is associated in the associated area, and they can communicate; however, how to ensure the suitability of the second customer and the first customer needs to be further resolved; this application first matches the attribute information of the two customers (for example: purpose of use, company size, etc.), matches the key attributes based on the function module-key attribute library, improves the matching efficiency, and completes the preliminary screening; then, the usage records of the two customers are matched, and key records are matched based on the function module-key record type library, which improves the matching efficiency; at the same time, based on the first time axis and the second time axis, the first value and the second value are cleverly determined, and the size of the second value is different depending on the selection direction, and the setting is reasonable, which improves the association efficiency.
[0187] An embodiment of the present invention provides a cloud platform customer management system, further comprising:
[0188] A monitoring module, used to monitor whether the first customer complies with the regulations when accessing the cloud platform, and if not, execute corresponding measures;
[0189] The monitoring module performs the following operations:
[0190] Periodically acquiring a plurality of first behavior items newly generated by the first customer;
[0191] Establishing a third timeline;
[0192] Based on a preset second setting rule, setting the first behavior item on the third time axis;
[0193] Extracting features from the first behavior items one by one from the starting point to the end point of the third time axis to obtain a plurality of seventh features;
[0194] Obtaining a preset trigger feature pair library, and randomly selecting a trigger feature pair from the trigger feature pair library, wherein the trigger feature pair includes: a first trigger feature and at least one second trigger feature that correspond to each other;
[0195] matching the seventh feature with the first trigger feature;
[0196] If the match is satisfied, determining the number of the second trigger features corresponding to the first trigger features that match the match;
[0197] Selecting the number of the first behavior items after the first behavior item for feature extraction on the third time axis as the second behavior items;
[0198] Performing feature extraction on the second behavior item to obtain a plurality of eighth features;
[0199] Matching the eighth feature with the second trigger feature corresponding to the first trigger feature that matches;
[0200] If the match is met, the second trigger feature that matches the match is used as the target feature, and at the same time, the sixth position of the second behavior item corresponding to the eighth feature that matches the match on the third time axis is determined;
[0201] Query the preset trigger feature-acquisition direction-selection range-malicious behavior library to determine the acquisition direction, selection range and malicious behavior corresponding to the target feature;
[0202] Selecting the first behavior item within the selected range in the acquisition direction of the sixth position on the third time axis as the third behavior item;
[0203] If the third behavior item includes the malicious behavior, determining that the first client's access is unqualified;
[0204] Determine the measures corresponding to the malicious behavior based on a preset malicious behavior-measure library;
[0205] Carry out the measures described.
[0206] The working principle and beneficial effects of the above technical solution are:
[0207] The preset second setting rule is specifically: the generation time of the behavior item is set in one-to-one correspondence with the time node on the time axis; the preset trigger feature pair library is specifically: a database containing multiple trigger feature pairs, and the trigger feature pair includes: a first trigger feature and at least one second trigger feature corresponding to the first trigger feature; the preset trigger feature-acquisition direction-selection range-malicious behavior library is specifically: containing the acquisition direction, selection range and malicious behavior corresponding to different second trigger features;
[0208] For example: a malicious behavior is divided into step 1, step 2 and step 3 according to the time sequence (for example: intrusion into the background, verbal abuse in the chat area, etc.). The first trigger feature can be the feature of step 1, and the second trigger feature can be the feature of step 2. The acquisition direction is to select backward on the third time axis, and the selection range can be determined based on the average required time between step 2 and step 3. If the selected third behavior item includes step 3, the malicious behavior is confirmed; the first trigger feature can also be the feature of step 2, and the second trigger feature is the feature of step 3. The acquisition direction is to select forward on the third time axis, and the selection range can be determined based on the average required time between step 1 and step 3. If the selected third behavior item includes step 1, the malicious behavior is confirmed;
[0209] The embodiment of the present invention matches the extracted seventh feature with the first trigger feature based on the trigger feature library. If the match is confirmed, it means that the first customer has the possibility of engaging in malicious behavior and the process continues; otherwise, the process does not continue, thereby improving the efficiency of malicious behavior monitoring. At the same time, based on the trigger feature-acquisition direction-selection range-malicious behavior library, the second trigger feature is used to further determine whether the malicious behavior is confirmed, and the acquisition direction and selection range can be directly determined, thereby improving the efficiency of confirming the confirmation.
[0210] An embodiment of the present invention provides a cloud platform customer management system, further comprising:
[0211] A rejection module, used for obtaining a preset malicious record library, and determining whether to reject the first customer based on the malicious record library, and if so, rejecting the first customer;
[0212] The removal module performs the following operations:
[0213] Determine at least one sixth record item corresponding to the first customer from the malicious record library;
[0214] Extracting a first impact value and a first dynamic impact value in the sixth record item;
[0215] Based on the first impact value and the first dynamic impact value, a first determination index is calculated, and the calculation formula is as follows:
[0216]
[0217] Among them, σ 1 is the first determination index, α 1,t is the first impact value in the tth sixth record item corresponding to the first customer, β 1,t is the first dynamic impact value in the tth sixth record item corresponding to the first customer, and O is the total number of sixth record items corresponding to the first customer;
[0218] Based on a preset customer-associated customer database, determining a plurality of associated customers corresponding to the first customer;
[0219] Acquire the association relationship between the associated customer and the customer;
[0220] Parsing the association relationship to obtain a first relationship value;
[0221] Obtaining a preset association prediction model, inputting the association relationship into the association prediction model, and obtaining a second relationship value;
[0222] Determine at least one seventh record item corresponding to the associated customer from the malicious record library;
[0223] Extracting the second impact value and the second dynamic impact value in the seventh record item;
[0224] Based on the first relationship value, the second relationship value, the second influence value and the second dynamic influence value, the second determination index is calculated as follows:
[0225]
[0226] Among them, σ 2 is the second determination index, A γ is the first relationship value corresponding to the γth associated customer corresponding to the first customer, B γ is the second relationship value corresponding to the γth associated customer corresponding to the first customer, n is the total number of associated customers corresponding to the first customer, α 2,γ,t is the second impact value of the t-th seventh record item corresponding to the γ-th associated customer corresponding to the first customer, β 2,γ,t is the second dynamic impact value in the t-th seventh record item corresponding to the γ-th associated customer corresponding to the first customer, l γ is the total number of seventh record items corresponding to the γth associated customers corresponding to the first customer;
[0227] If the first determination index is greater than or equal to a preset first determination index threshold and / or the second determination index is greater than or equal to a preset second determination index threshold, the corresponding first customer is eliminated.
[0228] The working principle and beneficial effects of the above technical solution are:
[0229] The preset malicious record library is specifically: a database containing historical malicious records of different customers (for example: malicious intrusion, chat abuse, etc.); the preset customer-associated customer library is specifically: a database containing associated customers corresponding to different customers, and customers can set associated customers when registering; the preset association prediction model is specifically: a model generated by learning a large number of records of manually predicted association closeness using a machine learning algorithm, and the model can predict the closeness of the association in the future and obtain a second relationship value. The larger the second relationship value, the higher the closeness; the preset first judgment index threshold is specifically: for example, 75; the preset second judgment index threshold is specifically: for example, 80;
[0230] When the first customer generates a malicious record, the impact of the malicious record generated at that time is immediately determined manually to obtain a first impact value. At the same time, the impact of the malicious record generated later is continuously determined and accumulated to obtain a first dynamic impact value. The second impact value and the second dynamic impact value are the same.
[0231] Calculate the first judgment index. When the first judgment index is greater than or equal to the first judgment index threshold, it means that the first customer's behavior is not good. In the corresponding formula, α 1,t The larger the β 1,t , the larger the first judgment index, the greater the 1,t -α 1,t The larger the value, the greater the impact in the later stage, and the larger the first judgment index; when calculating the second judgment index, if the second judgment index is greater than or equal to the second judgment index threshold, it means that the behavior of the first customer's associated customers is not good; in the corresponding formula, α 2,γ,t The larger the β 2,γ,t The larger the value, the larger the second judgment index, β 2,γ,t -α 2,γ,t The larger the value, the greater the later impact, the larger the second judgment index, and the γ +B γ The larger the B γ -A γ The larger the value, the closer the relationship, the greater the guarantee relationship, and the larger the second determination index, the smaller the range of the second impact value and the second dynamic impact value that can be increased; the first customer is eliminated and is not allowed to continue to use the cloud platform;
[0232] A guarantee relationship is formed between the first customer and the associated customers, that is, if one party generates malicious records, the other party will also be affected, which increases the cost for them to generate malicious records, inhibits the occurrence of malicious behavior to a certain extent, improves the stability of platform use, and maintains platform order.
[0233] The embodiment of the present invention provides a cloud platform customer management method, such as Figure 2 As shown, including:
[0234] Step S1: Obtain order information of a first customer, and determine evaluation criteria based on the order information;
[0235] Step S2: when the duration of the first customer's use of the cloud platform reaches a preset duration threshold, obtaining a first usage record of the first customer, and evaluating the first usage record based on the evaluation criteria;
[0236] Step S3: Generate recommended content based on the evaluation results and recommend it to the first customer.
[0237] The working principle and beneficial effects of the above technical solution are:
[0238] Obtain the order information of the first customer (for example, an 8-core CPU is ordered for a period of 1 month); determine the evaluation criteria based on the order information (for example, whether the computing power of the 8-core CPU can meet the usage requirements of the first customer); when the usage time of the first customer using the cloud platform reaches a preset time threshold (for example, 20 days), obtain the usage record of the first customer, and evaluate the usage record based on the evaluation criteria; generate recommendation content (recommend upgrading the CPU to 12 cores) based on the evaluation result (for example, the computing power of the 8-core CPU cannot meet the usage requirements of the first customer);
[0239] The embodiment of the present invention can evaluate the first usage record of the first customer, and based on the evaluation result, accurately recommend to the customer how to choose the paid function on the cloud platform, thereby improving the customer experience and being very user-friendly.
[0240] The embodiment of the present invention provides a cloud platform customer management method, such as Figure 3 As shown, in step S1, based on the order information, determining the evaluation criteria includes:
[0241] Step S101: extracting multiple order items in the order information, wherein the order items include: order type and order duration;
[0242] Step S102: determining the evaluation criteria corresponding to the subscription type and the subscription duration based on a preset subscription type-subscription duration-evaluation criteria library;
[0243] Step S103: Integrate the evaluation criteria to complete the determination.
[0244] The working principle and beneficial effects of the above technical solution are:
[0245] The preset order type-order duration-evaluation standard library is specifically: a database containing evaluation standards corresponding to different order types and different order durations. For example, if the order type is to order an 8-core CPU and the order duration is 15 days, the corresponding evaluation standard is whether the average level of the 8-core CPU within 15 days meets the customer's business processing requirements;
[0246] Based on the order type, order duration and evaluation standard library, determine the evaluation standards, and complete the determination by integration.
[0247] An embodiment of the present invention provides a cloud platform customer management method, wherein in step S2, the first usage record is evaluated based on the evaluation criteria, including:
[0248] extracting a target record corresponding to the subscription type from the first usage record;
[0249] Based on the evaluation criteria, the corresponding target records are evaluated to obtain evaluation results;
[0250] Integrate the evaluation results to complete the evaluation.
[0251] The working principle and beneficial effects of the above technical solution are:
[0252] For example, if the order type is to order an 8-core CPU, the target record corresponding to the order type in the first usage record is extracted, that is, the load condition (for example, overload) of the 8-core CPU when the customer performs business processing;
[0253] Based on the evaluation criteria, the target records are evaluated to complete the evaluation.
[0254] The embodiment of the present invention provides a cloud platform customer management method, wherein in step S3, based on the evaluation result, recommendation content is generated and recommended to the first customer, including:
[0255] Extracting multiple evaluation items from the evaluation results, wherein the evaluation items include: evaluation type and score;
[0256] Determine the recommended content corresponding to the evaluation type and the score based on a preset evaluation type-score-recommended content library;
[0257] The recommended contents are integrated to complete the generation, and at the same time, are recommended to the first customer.
[0258] The working principle and beneficial effects of the above technical solution are:
[0259] The preset evaluation type-score-recommendation content library is specifically: including recommended content corresponding to different evaluation types and different scores. For example: the evaluation type is that the CPU cannot meet the requirements, and the score is 10 (the lower the score, the more serious the situation where the CPU cannot meet the requirements), and the corresponding recommended content is that it is recommended to upgrade the CPU to 12 cores.
[0260] An embodiment of the present invention provides a cloud platform customer management method, further comprising:
[0261] Based on the evaluation result, the first customer is compensated accordingly;
[0262] Wherein, based on the evaluation result, the first customer is compensated accordingly, including:
[0263] Based on a preset evaluation type-score-compensation method library, try to determine a compensation method corresponding to the evaluation type and the score;
[0264] Based on the compensation method, the first customer is compensated.
[0265] The working principle and beneficial effects of the above technical solution are:
[0266] The preset evaluation type-score-compensation method library specifically includes compensation methods corresponding to different evaluation types and different scores. For example, if the evaluation type is that the CPU cannot meet the demand and the score is 10, the corresponding compensation method is a free trial of a 12-core CPU.
[0267] An embodiment of the present invention provides a cloud platform customer management method, further comprising:
[0268] When the first customer accesses the function module of the cloud platform, an association area is constructed, and a suitable second customer is associated with the first customer in the association area, so that the first customer and the second customer can communicate in the association area;
[0269] Wherein, associating a suitable second customer with the first customer in the association area includes:
[0270] Acquire a preset online customer set, the online customer set comprising: a plurality of third customers;
[0271] confirming whether the third client is accessing the functional module;
[0272] If so, the corresponding third customer is regarded as the fourth customer;
[0273] respectively obtaining first attribute information of the first customer and second attribute information of the fourth customer;
[0274] Based on a preset function module-key attribute library, determining at least one key attribute corresponding to the function module;
[0275] Determining a plurality of first attribute items corresponding to the key attribute in the first attribute information;
[0276] Determining a plurality of second attribute items corresponding to the key attribute in the second attribute information;
[0277] Performing feature extraction on the first attribute item to obtain a plurality of first features;
[0278] Performing feature extraction on the second attribute item to obtain a plurality of second features;
[0279] Matching the first feature with the second feature to obtain a first matching degree;
[0280] Summarizing the first matching degrees to obtain a first matching degree sum, and matching it with the corresponding key type;
[0281] Based on a preset key attribute-matching degree-association determination value library, determining a first association determination value corresponding to the key attribute and the first matching degree;
[0282] Summarizing the first association determination values to obtain a sum of association determination values;
[0283] If the sum of the association determination values is greater than or equal to a preset first threshold, the fourth customer is regarded as the fifth customer;
[0284] Obtaining a second usage record of the fifth customer;
[0285] Based on a preset function module-key record type library, determining at least one key record type corresponding to the function module;
[0286] determining a plurality of first record items corresponding to the key record type in the first usage record;
[0287] determining a plurality of second record items corresponding to the key record type in the second usage record;
[0288] Establish a first time axis and a second time axis respectively;
[0289] Based on a preset first setting rule, the first record item is set on the first time axis, and at the same time, the second record item is set on the second time axis;
[0290] Randomly extracting features from one of the first record items to obtain a plurality of third features;
[0291] Determining a first position of the first record item for feature extraction on the first time axis;
[0292] Determine the second record item within a preset first range before and / or after a second position corresponding to the first position on the second time axis, and use it as the third record item;
[0293] Performing feature extraction on the third record item to obtain a plurality of fourth features;
[0294] Matching the third feature with the fourth feature to obtain a second matching degree;
[0295] Summarizing the second matching degrees to obtain a second matching degree sum, and matching it with the corresponding key record type;
[0296] Based on the preset key record type-matching degree and-value degree, determining the key record type and the second matching degree and the first value degree corresponding to each other;
[0297] If the first value is greater than or equal to a preset second threshold, determining a third position corresponding to the third record item on the second time axis;
[0298] Setting a selection direction, the selection direction includes: front or back;
[0299] Selecting the first record item in the selection direction of the first record item for feature extraction on the first time axis as the fourth record item;
[0300] Determining a fourth position of the fourth record item on the first time axis;
[0301] Determine the second record item within a preset second range before and / or after a fifth position corresponding to the fourth position on the second time axis, and use the second record item as the fifth record item;
[0302] Performing feature extraction on the fourth record item to obtain a plurality of fifth features;
[0303] Performing feature extraction on the fifth record item to obtain a plurality of sixth features;
[0304] Matching the fifth feature with the sixth feature to obtain a third matching degree;
[0305] Summarizing the third matching degrees to obtain a third matching degree sum;
[0306] Based on a preset key record type-selection direction-matching degree and-value degree library, determine the key record type, selection direction and third matching degree and the corresponding second value degree;
[0307] Summarize the first value and the second value to obtain a second association determination value, and correspond it to the corresponding key record type;
[0308] Summarizing the second association determination values to obtain a second association determination value sum;
[0309] If the sum of the second association determination value is greater than or equal to a preset third threshold, the corresponding fifth customer will be regarded as the second customer.
[0310] The working principle and beneficial effects of the above technical solution are:
[0311] The preset online customer set is specifically: a collection of multiple online customers (third customers); the preset function module-key attribute library is specifically: a database containing key attributes corresponding to different function modules, for example: the function module is a storage function area (providing cloud storage function), the key attribute is the customer's usage scenario (used to store department information, the number of department staff is 10), etc.; the preset key attribute-matching degree and-association judgment value library is specifically: containing different key attributes and different matching degrees and corresponding association judgment values, the larger the matching degree and the larger the association judgment value, for example: the key attribute is the customer's usage scenario, the first matching degree and the corresponding association judgment value is 87; the preset first threshold is specifically: for example, 820; the preset function module-key record type library is specifically: a database containing key record types corresponding to different function modules, for example: the function module is a storage function area, and the key record type is daily upload volume (the amount of data uploaded by the customer to the cloud storage area every day); the preset first setting rule is specifically: the generation time of the record item and the time axis are The time nodes are set one by one; the preset first range is specifically: the range value on the time axis, for example: 0.5 days; the preset key record type-matching degree and-value are specifically: a database containing different key record types, different matching degrees and corresponding first values. The greater the matching degree, the greater the value. For example: the key record type is daily upload volume, the second matching degree is 50, and the first value is 66; the preset second threshold is specifically: for example, 750; the preset second range is specifically: the range value on the time axis, for example: 0.6 days; the preset key record type-selection direction-matching degree and-value database is specifically: containing different key record types, different selection directions and different third matching degrees and corresponding values. Different selection directions also affect the value. For example: the usage record is daily upload volume. The more the daily upload volume of the first customer and the second customer (the selection direction is later) matches, the more their usage is similar, the higher the value, and the previous daily upload volume has a lower value; the preset third threshold is specifically: for example, 800;
[0312] When the cloud platform is actually used, even if it recommends to customers how to choose paid functions, since customers all hope to save costs, how to use the functions more reasonably is more important. Due to the influence of customers, the platform cannot obtain customer data, but customers can communicate with each other. For example, the first customer and the second customer are both financial directors of companies of similar size, and store invoices, payrolls, etc. on the cloud platform. However, how to compress the invoices to ensure that the size of the data is minimized before uploading, the first customer and the second customer communicate to better and reasonably use the storage function of the cloud platform; therefore, this application constructs an association area (a functional submodule that can be displayed as a floating window when the customer accesses the functional module) for the first customer The second customer is associated in the associated area, and they can communicate; however, how to ensure the suitability of the second customer and the first customer needs to be further resolved; this application first matches the attribute information of the two customers (for example: purpose of use, company size, etc.), matches the key attributes based on the function module-key attribute library, improves the matching efficiency, and completes the preliminary screening; then, the usage records of the two customers are matched, and key records are matched based on the function module-key record type library, which improves the matching efficiency; at the same time, based on the first time axis and the second time axis, the first value and the second value are cleverly determined, and the size of the second value is different depending on the selection direction, and the setting is reasonable, which improves the association efficiency.
[0313] An embodiment of the present invention provides a cloud platform customer management method, further comprising:
[0314] monitoring whether the first customer complies with the regulations when accessing the cloud platform, and if not, executing corresponding measures;
[0315] Wherein, monitoring whether the first customer is in compliance when accessing the cloud platform, and if not, executing corresponding measures, including:
[0316] Periodically acquiring a plurality of first behavior items newly generated by the first customer;
[0317] Establishing a third timeline;
[0318] Based on a preset second setting rule, setting the first behavior item on the third time axis;
[0319] Extracting features from the first behavior items one by one from the starting point to the end point of the third time axis to obtain a plurality of seventh features;
[0320] Obtaining a preset trigger feature pair library, and randomly selecting a trigger feature pair from the trigger feature pair library, wherein the trigger feature pair includes: a first trigger feature and at least one second trigger feature that correspond to each other;
[0321] matching the seventh feature with the first trigger feature;
[0322] If the match is satisfied, determining the number of the second trigger features corresponding to the first trigger features that match the match;
[0323] Selecting the number of the first behavior items after the first behavior item for feature extraction on the third time axis as the second behavior items;
[0324] Performing feature extraction on the second behavior item to obtain a plurality of eighth features;
[0325] Matching the eighth feature with the second trigger feature corresponding to the first trigger feature that matches;
[0326] If the match is met, the second trigger feature that matches the match is used as the target feature, and at the same time, the sixth position of the second behavior item corresponding to the eighth feature that matches the match on the third time axis is determined;
[0327] Query the preset trigger feature-acquisition direction-selection range-malicious behavior library to determine the acquisition direction, selection range and malicious behavior corresponding to the target feature;
[0328] Selecting the first behavior item within the selected range in the acquisition direction of the sixth position on the third time axis as the third behavior item;
[0329] If the third behavior item includes the malicious behavior, determining that the first client's access is unqualified;
[0330] Determine the measures corresponding to the malicious behavior based on a preset malicious behavior-measure library;
[0331] Carry out the measures described.
[0332] The working principle and beneficial effects of the above technical solution are:
[0333] The preset second setting rule is specifically: the generation time of the behavior item is set in one-to-one correspondence with the time node on the time axis; the preset trigger feature pair library is specifically: a database containing multiple trigger feature pairs, and the trigger feature pair includes: a first trigger feature and at least one second trigger feature corresponding to the first trigger feature; the preset trigger feature-acquisition direction-selection range-malicious behavior library is specifically: containing the acquisition direction, selection range and malicious behavior corresponding to different second trigger features;
[0334] For example: a malicious behavior is divided into step 1, step 2 and step 3 according to the time sequence (for example: intrusion into the background, verbal abuse in the chat area, etc.). The first trigger feature can be the feature of step 1, and the second trigger feature can be the feature of step 2. The acquisition direction is to select backward on the third time axis, and the selection range can be determined based on the average required time between step 2 and step 3. If the selected third behavior item includes step 3, the malicious behavior is confirmed; the first trigger feature can also be the feature of step 2, and the second trigger feature is the feature of step 3. The acquisition direction is to select forward on the third time axis, and the selection range can be determined based on the average required time between step 1 and step 3. If the selected third behavior item includes step 1, the malicious behavior is confirmed;
[0335] The embodiment of the present invention matches the extracted seventh feature with the first trigger feature based on the trigger feature library. If the match is confirmed, it means that the first customer has the possibility of engaging in malicious behavior and the process continues; otherwise, the process does not continue, thereby improving the efficiency of malicious behavior monitoring. At the same time, based on the trigger feature-acquisition direction-selection range-malicious behavior library, the second trigger feature is used to further determine whether the malicious behavior is confirmed, and the acquisition direction and selection range can be directly determined, thereby improving the efficiency of confirming the confirmation.
[0336] An embodiment of the present invention provides a cloud platform customer management method, further comprising:
[0337] Obtain a preset malicious record library, and determine whether to remove the first customer based on the malicious record library, and if so, remove the first customer;
[0338] Wherein, based on the malicious record library, determining whether to remove the first customer, and if so, removing the first customer comprises:
[0339] Determine at least one sixth record item corresponding to the first customer from the malicious record library;
[0340] Extracting a first impact value and a first dynamic impact value in the sixth record item;
[0341] Based on the first impact value and the first dynamic impact value, a first determination index is calculated, and the calculation formula is as follows:
[0342]
[0343] Among them, σ 1 is the first determination index, α 1,t is the first impact value in the tth sixth record item corresponding to the first customer, β 1,t is the first dynamic impact value in the tth sixth record item corresponding to the first customer, and O is the total number of sixth record items corresponding to the first customer;
[0344] Based on a preset customer-associated customer database, determining a plurality of associated customers corresponding to the first customer;
[0345] Acquire the association relationship between the associated customer and the customer;
[0346] Parsing the association relationship to obtain a first relationship value;
[0347] Obtaining a preset association prediction model, inputting the association relationship into the association prediction model, and obtaining a second relationship value;
[0348] Determine at least one seventh record item corresponding to the associated customer from the malicious record library;
[0349] Extracting the second impact value and the second dynamic impact value in the seventh record item;
[0350] Based on the first relationship value, the second relationship value, the second influence value and the second dynamic influence value, the second determination index is calculated as follows:
[0351]
[0352] Among them, σ 2 is the second determination index, A γ is the first relationship value corresponding to the γth associated customer corresponding to the first customer, B γ is the second relationship value corresponding to the γth associated customer corresponding to the first customer, n is the total number of associated customers corresponding to the first customer, α 2,γ,t is the second impact value of the t-th seventh record item corresponding to the γ-th associated customer corresponding to the first customer, β 2,γ,t is the second dynamic impact value in the t-th seventh record item corresponding to the γ-th associated customer corresponding to the first customer, l γ is the total number of seventh record items corresponding to the γth associated customers corresponding to the first customer;
[0353] If the first determination index is greater than or equal to a preset first determination index threshold and / or the second determination index is greater than or equal to a preset second determination index threshold, the corresponding first customer is eliminated.
[0354] The working principle and beneficial effects of the above technical solution are:
[0355] The preset malicious record library is specifically: a database containing historical malicious records of different customers (for example: malicious intrusion, chat abuse, etc.); the preset customer-associated customer library is specifically: a database containing associated customers corresponding to different customers, and customers can set associated customers when registering; the preset association prediction model is specifically: a model generated by learning a large number of records of manually predicted association closeness using a machine learning algorithm, and the model can predict the closeness of the association in the future and obtain a second relationship value. The larger the second relationship value, the higher the closeness; the preset first judgment index threshold is specifically: for example, 75; the preset second judgment index threshold is specifically: for example, 80;
[0356] When the first customer generates a malicious record, the impact of the malicious record generated at that time is immediately determined manually to obtain a first impact value. At the same time, the impact of the malicious record generated later is continuously determined and accumulated to obtain a first dynamic impact value. The second impact value and the second dynamic impact value are the same.
[0357] Calculate the first judgment index. When the first judgment index is greater than or equal to the first judgment index threshold, it means that the first customer's behavior is not good. In the corresponding formula, α 1,t The larger the β 1,t , the larger the first judgment index, the greater the 1,t -α 1,t The larger the value, the greater the impact in the later stage, and the larger the first judgment index; when calculating the second judgment index, if the second judgment index is greater than or equal to the second judgment index threshold, it means that the behavior of the first customer's associated customers is not good; in the corresponding formula, α 2,γ,t The larger the β 2,γ,t The larger the value, the larger the second judgment index, β 2,γ,t -α 2,γ,t The larger the value, the greater the later impact, the larger the second judgment index, and the γ +B γ The larger the B γ -A γ The larger the value, the closer the relationship, the greater the guarantee relationship, and the larger the second determination index, the smaller the range of the second impact value and the second dynamic impact value that can be increased; the first customer is eliminated and is not allowed to continue to use the cloud platform;
[0358] A guarantee relationship is formed between the first customer and the associated customers, that is, if one party generates malicious records, the other party will also be affected, which increases the cost for them to generate malicious records, inhibits the occurrence of malicious behavior to a certain extent, improves the stability of platform use, and maintains platform order.
[0359] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.
Claims
1. A cloud platform customer management system, characterized in that: include: A determination module, configured to obtain order information of a first customer and determine an evaluation criterion based on the order information; an evaluation module, configured to obtain a first usage record of the first customer when the duration of use of the cloud platform by the first customer reaches a preset duration threshold, and evaluate the first usage record based on the evaluation criteria; A recommendation module, used to generate recommendation content based on the evaluation result and recommend it to the first customer; A monitoring module, used to monitor whether the first customer complies with the regulations when accessing the cloud platform, and if not, execute corresponding measures; The monitoring module performs the following operations: Periodically acquiring a plurality of first behavior items newly generated by the first customer; Establishing a third timeline; Based on a preset second setting rule, setting the first behavior item on the third time axis; Extracting features from the first behavior items one by one from the starting point to the end point of the third time axis to obtain a plurality of seventh features; Obtaining a preset trigger feature pair library, and randomly selecting a trigger feature pair from the trigger feature pair library, wherein the trigger feature pair includes: a first trigger feature and at least one second trigger feature that correspond to each other; matching the seventh feature with the first trigger feature; If the match is satisfied, determining the number of the second trigger features corresponding to the first trigger features that match the match; Selecting the number of the first behavior items after the first behavior item for feature extraction on the third time axis as the second behavior items; Performing feature extraction on the second behavior item to obtain a plurality of eighth features; Matching the eighth feature with the second trigger feature corresponding to the first trigger feature that matches; If the match is met, the second trigger feature that matches the match is used as the target feature, and at the same time, the sixth position of the second behavior item corresponding to the eighth feature that matches the match on the third time axis is determined; Query the preset trigger feature-acquisition direction-selection range-malicious behavior library to determine the acquisition direction, selection range and malicious behavior corresponding to the target feature; Selecting the first behavior item within the selected range in the acquisition direction of the sixth position on the third time axis as the third behavior item; If the third behavior item includes the malicious behavior, determining that the first client's access is unqualified; Determine the measures corresponding to the malicious behavior based on a preset malicious behavior-measure library; Carry out the measures described.
2. A cloud platform customer management system according to claim 1, characterized in that: The determination module performs the following operations: Extracting multiple order items from the order information, the order items including: order type and order duration; Based on a preset subscription type-subscription duration-evaluation standard library, determining an evaluation standard corresponding to the subscription type and the subscription duration; Integrate the above-mentioned evaluation criteria and complete the determination.
3. A cloud platform customer management system as claimed in claim 2, characterized in that: The evaluation module performs the following operations: extracting a target record corresponding to the subscription type from the first usage record; Based on the evaluation criteria, the corresponding target records are evaluated to obtain evaluation results; Integrate the evaluation results to complete the evaluation.
4. A cloud platform customer management system according to claim 1, characterized in that: The recommendation module performs the following operations: Extracting multiple evaluation items from the evaluation results, wherein the evaluation items include: evaluation type and score; Determine the recommended content corresponding to the evaluation type and the score based on a preset evaluation type-score-recommended content library; The recommended contents are integrated to complete the generation, and at the same time, are recommended to the first customer.
5. A cloud platform customer management system according to claim 1, characterized in that: Also includes: A compensation module, used for compensating the first customer accordingly based on the evaluation result; The compensation module performs the following operations: Based on a preset evaluation type-score-compensation method library, try to determine a compensation method corresponding to the evaluation type and the score; Based on the compensation method, the first customer is compensated.
6. A cloud platform customer management system according to claim 1, characterized in that: Also includes: a construction module, configured to construct an association area when the first customer accesses the function module of the cloud platform, and associate an appropriate second customer with the first customer in the association area, so as to facilitate communication between the first customer and the second customer in the association area; The building block performs the following operations: Acquire a preset online customer set, the online customer set comprising: a plurality of third customers; confirming whether the third client is accessing the functional module; If so, the corresponding third customer is regarded as the fourth customer; respectively obtaining first attribute information of the first customer and second attribute information of the fourth customer; Based on a preset function module-key attribute library, determining at least one key attribute corresponding to the function module; Determining a plurality of first attribute items corresponding to the key attribute in the first attribute information; Determining a plurality of second attribute items corresponding to the key attribute in the second attribute information; Performing feature extraction on the first attribute item to obtain a plurality of first features; Performing feature extraction on the second attribute item to obtain a plurality of second features; Matching the first feature with the second feature to obtain a first matching degree; Summarizing the first matching degrees to obtain a first matching degree sum, and matching it with the corresponding key type; Based on a preset key attribute-matching degree-association determination value library, determining a first association determination value corresponding to the key attribute and the first matching degree; Summarizing the first association determination values to obtain a sum of association determination values; If the sum of the association determination values is greater than or equal to a preset first threshold, the fourth customer is regarded as the fifth customer; Obtaining a second usage record of the fifth customer; Based on a preset function module-key record type library, determining at least one key record type corresponding to the function module; determining a plurality of first record items corresponding to the key record type in the first usage record; determining a plurality of second record items corresponding to the key record type in the second usage record; Establish a first time axis and a second time axis respectively; Based on a preset first setting rule, the first record item is set on the first time axis, and at the same time, the second record item is set on the second time axis; Randomly extracting features from one of the first record items to obtain a plurality of third features; Determining a first position of the first record item for feature extraction on the first time axis; Determine the second record item within a preset first range before and / or after a second position corresponding to the first position on the second time axis, and use it as the third record item; Performing feature extraction on the third record item to obtain a plurality of fourth features; Matching the third feature with the fourth feature to obtain a second matching degree; Summarizing the second matching degrees to obtain a second matching degree sum, and matching it with the corresponding key record type; Based on the preset key record type-matching degree and-value degree, determining the key record type and the second matching degree and the first value degree corresponding to each other; If the first value is greater than or equal to a preset second threshold, determining a third position corresponding to the third record item on the second time axis; Setting a selection direction, the selection direction includes: front or back; Selecting the first record item in the selection direction of the first record item for feature extraction on the first time axis as the fourth record item; Determining a fourth position of the fourth record item on the first time axis; Determine the second record item within a preset second range before and / or after a fifth position corresponding to the fourth position on the second time axis, and use the second record item as the fifth record item; Performing feature extraction on the fourth record item to obtain a plurality of fifth features; Performing feature extraction on the fifth record item to obtain a plurality of sixth features; Matching the fifth feature with the sixth feature to obtain a third matching degree; Summarizing the third matching degrees to obtain a third matching degree sum; Based on a preset key record type-selection direction-matching degree and-value degree library, determine the key record type, selection direction and third matching degree and the corresponding second value degree; Summarize the first value and the second value to obtain a second association determination value, and correspond it to the corresponding key record type; Summarizing the second association determination values to obtain a second association determination value sum; If the sum of the second association determination value is greater than or equal to a preset third threshold, the corresponding fifth customer will be regarded as the second customer.
7. A cloud platform customer management method, characterized in that: include: Step S1: Obtain order information of a first customer, and determine evaluation criteria based on the order information; Step S2: when the duration of the first customer's use of the cloud platform reaches a preset duration threshold, obtaining a first usage record of the first customer, and evaluating the first usage record based on the evaluation criteria; Step S3: generating recommended content based on the evaluation results, and recommending it to the first customer; A monitoring module, used to monitor whether the first customer complies with the regulations when accessing the cloud platform, and if not, execute corresponding measures; The monitoring module performs the following operations: Periodically acquiring a plurality of first behavior items newly generated by the first customer; Establishing a third timeline; Based on a preset second setting rule, setting the first behavior item on the third time axis; Extracting features from the first behavior items one by one from the starting point to the end point of the third time axis to obtain a plurality of seventh features; Obtaining a preset trigger feature pair library, and randomly selecting a trigger feature pair from the trigger feature pair library, wherein the trigger feature pair includes: a first trigger feature and at least one second trigger feature that correspond to each other; matching the seventh feature with the first trigger feature; If the match is satisfied, determining the number of the second trigger features corresponding to the first trigger features that match the match; Selecting the number of the first behavior items after the first behavior item for feature extraction on the third time axis as the second behavior items; Performing feature extraction on the second behavior item to obtain a plurality of eighth features; Matching the eighth feature with the second trigger feature corresponding to the first trigger feature that matches; If the match is met, the second trigger feature that matches the match is used as the target feature, and at the same time, the sixth position of the second behavior item corresponding to the eighth feature that matches the match on the third time axis is determined; Query the preset trigger feature-acquisition direction-selection range-malicious behavior library to determine the acquisition direction, selection range and malicious behavior corresponding to the target feature; Selecting the first behavior item within the selected range in the acquisition direction of the sixth position on the third time axis as the third behavior item; If the third behavior item includes the malicious behavior, determining that the first client's access is unqualified; Determine the measures corresponding to the malicious behavior based on a preset malicious behavior-measure library; Carry out the measures described.
8. A cloud platform customer management method as claimed in claim 7, characterized in that: In step S1, based on the order information, determining the evaluation criteria includes: Extracting multiple order items from the order information, the order items including: order type and order duration; Based on a preset subscription type-subscription duration-evaluation standard library, determining an evaluation standard corresponding to the subscription type and the subscription duration; Integrate the above-mentioned evaluation criteria and complete the determination.
9. A cloud platform customer management method as claimed in claim 8, characterized in that: In the step S2, based on the evaluation criteria, the first usage record is evaluated, including: extracting a target record corresponding to the subscription type from the first usage record; Based on the evaluation criteria, the corresponding target records are evaluated to obtain evaluation results; Integrate the evaluation results to complete the evaluation.
