Employee experience optimization method and system based on cloud computing and low code

Abstract

The invention relates to the technical field of cloud computing, and discloses an employee experience optimization method and system based on cloud computing and low code, and the method comprises the steps: obtaining employee behavior data, a task key weight, a task safety level and a resource distribution state; performing emotion analysis according to the employee behavior data to obtain employee emotion data; performing priority scheduling according to the employee emotion data and the task key weight to obtain a priority list; performing permission adaptability evaluation according to the employee emotion data and the task security level to obtain a permission change report; performing resource scheduling according to the employee emotion data and the resource allocation state to obtain a resource scheduling report; and generating a task sequence according to the priority list, the permission change report and the resource scheduling report, and outputting a final task planning table. The method has the following effect that the employee experience can be optimized.

Description

Technical Field

[0001] The present invention relates to the field of cloud computing technology, and in particular to an employee experience optimization method and system based on cloud computing and low code. Background Art

[0002] In today's era of digital transformation, employee experience optimization methods and systems based on cloud computing and low-code platforms have gradually become the key to improving the competitiveness of enterprises. These systems integrate multiple functional modules, such as task management, permission control, resource scheduling, etc., to improve work efficiency and optimize employees' work experience. Cloud computing provides powerful computing power and flexible resource allocation mechanisms, while low-code platforms enable non-technical personnel to quickly build applications and simplify business processes. However, in actual applications, these technical solutions often focus on technical optimization, while ignoring the impact of employee emotions, an important human factor, on work performance.

[0003] In one existing technology, the existing employee experience optimization method based on cloud computing and low code includes the following steps: First, the system automatically analyzes data such as employee task completion, project progress, and resource utilization efficiency; second, based on these data analysis results, the system dynamically adjusts task allocation to ensure that each employee can get a work task that suits their skill level; in addition, the system will intelligently recommend corresponding permission settings based on the employee's role and responsibilities to ensure information security while improving work flexibility; finally, based on real-time monitoring data, the system will automatically adjust computing resources to ensure stable operation and efficient use of the system. This method relies on algorithms and data analysis to achieve automated decision-making, strive to reduce human intervention, and improve overall work efficiency.

[0004] Although the above methods have improved work efficiency and resource utilization to a certain extent, their main disadvantage is that they do not fully consider the emotional state of employees and make timely adjustments to task difficulty, task authority, resource scheduling, etc. based on this. The derivation process is as follows: First, the emotional state of employees directly affects their work efficiency and creativity. If the system cannot perceive the emotional changes of employees (such as feeling too stressed or depressed), it will not be able to adjust the task difficulty accordingly, which will cause employees to suffer from excessive pressure and affect their health or work quality. Secondly, in terms of authority management, employees' emotional fluctuations will also affect their demand for information and processing capabilities, but the existing system does not take this into account, resulting in inflexible authority settings and failure to meet the needs of employees in different emotional states. Finally, in resource scheduling, ignoring employee emotions will cause waste or shortage of resources, such as when employees need more support but fail to get appropriate resources. In summary, although the existing methods have achieved optimization at the technical and operational levels, they still have obvious limitations in dealing with the challenges brought by employee emotional changes, resulting in poor employee experience. Summary of the invention

[0005] The present invention provides an employee experience optimization method and system based on cloud computing and low-code to achieve the goal of optimizing employee work experience.

[0006] In the first aspect, in order to solve the above technical problems, the present invention provides an employee experience optimization method based on cloud computing and low code, comprising:

[0007] Obtain employee behavior data, task criticality weights, task safety levels, and resource allocation status;

[0008] Performing sentiment analysis based on the employee behavior data to obtain employee sentiment data;

[0009] Performing priority scheduling according to the employee emotion data and the task criticality weights to obtain a priority list;

[0010] Perform authority adaptability assessment based on the employee emotion data and the task security level to obtain an authority change report;

[0011] Perform resource scheduling according to the employee emotion data and the resource allocation status to obtain a resource scheduling report;

[0012] A task sequence is generated according to the priority list, the authority change report and the resource scheduling report, and a final task planning table is output.

[0013] In an optional implementation, performing emotion analysis based on the employee behavior data to obtain employee emotion data includes:

[0014] Extracting text from the voice records in the employee behavior data to obtain voice text;

[0015] Performing emotion recognition based on the voice text and the chat text of the employee behavior data to obtain an emotion state label;

[0016] Constructing an emotional state sequence according to the emotional state labels of different time periods;

[0017] Calculating the standard deviation of the emotional state sequence to obtain the emotional fluctuation amplitude;

[0018] Performing frequency statistics on the emotional state sequence to obtain the frequency of emotional fluctuations;

[0019] The employee emotion data includes the emotion fluctuation amplitude and the emotion fluctuation frequency.

[0020] In an optional implementation, the priority scheduling is performed according to the employee emotion data and the task criticality weight to obtain a priority list, including:

[0021] According to the task criticality weights, the to-do tasks corresponding to the task criticality weights are sorted in descending order to generate an initial task sequence;

[0022] When the emotion fluctuation amplitude in the employee emotion data is greater than a preset amplitude threshold, all long-term planning tasks in the initial task sequence are removed to obtain a short-term task sequence;

[0023] Balance adjustments are made according to the short-term task sequence to obtain a priority list.

[0024] In an optional implementation, the authority adaptability assessment is performed according to the employee emotion data and the task security level to obtain an authority change report, including:

[0025] Get historical permission records;

[0026] Calculate the authority threshold according to the historical authority record and the task security level to obtain the authority threshold;

[0027] When the frequency of emotional fluctuations in the employee's emotional data is greater than the authority threshold, the task authority corresponding to the task security level is adjusted to be inaccessible;

[0028] When the frequency of the emotion fluctuation is less than the permission threshold, adjusting the task permission to be accessible;

[0029] The permission change report includes the task permission and the corresponding change record.

[0030] In an optional implementation, the calculating the authority threshold according to the historical authority record and the task security level to obtain the authority threshold includes:

[0031] The permission threshold is calculated using the following formula:

[0032]

[0033] in, express The permission threshold at the moment, Indicates the current moment, Represents the historical permission baseline, represents the historical authority coefficient, express The real-time entropy weight at the moment, Indicates the security level is The safety level adjustment factor is Indicates the task safety level, represents the coefficient of deviation, represents the natural base, Indicates the absolute deviation of the current authority value from the historical mean.

[0034] In an optional implementation, performing resource scheduling according to the employee emotion data and the resource allocation status to obtain a resource scheduling report includes:

[0035] Performing idle computing according to the resource allocation state to obtain idle computing resources;

[0036] When the frequency of emotion fluctuations in the employee emotion data is greater than a preset frequency threshold, allocating the idle computing resources to the employee;

[0037] When the emotion fluctuation data is less than a preset frequency threshold, resource recovery calculation is performed based on the employee emotion data to obtain a resource recovery ratio;

[0038] Reclaiming computing resources allocated to employees according to the resource recovery ratio, and updating the resource allocation state;

[0039] The resource scheduling report includes resource scheduling records and resource allocation status.

[0040] In an optional implementation, generating a task sequence according to the priority list, the authority change report and the resource scheduling report, and outputting a final task planning table includes:

[0041] Performing permission matching according to the priority list and the permission change report to obtain a permission adaptation task sequence;

[0042] Perform sorting optimization according to the resource scheduling report and the permission adaptation task sequence to generate a preliminary execution table;

[0043] Get historical execution deviation data;

[0044] Fault tolerance correction is performed based on the historical execution deviation data, and a final task planning table is output.

[0045] In a second aspect, the present invention provides an employee experience optimization system based on cloud computing and low code, comprising:

[0046] Data acquisition module, used to obtain employee behavior data, task criticality weight, task safety level and resource allocation status;

[0047] A sentiment analysis module, used to perform sentiment analysis based on the employee behavior data to obtain employee sentiment data;

[0048] A weight scheduling module, used to perform priority scheduling according to the employee emotion data and the task criticality weight to obtain a priority list;

[0049] The permission change module is used to evaluate the suitability of permissions based on the employee emotion data and the task security level to obtain a permission change report;

[0050] A resource scheduling module, used to perform resource scheduling according to the employee emotion data and the resource allocation status, and obtain a resource scheduling report;

[0051] The task planning module is used to generate a task sequence according to the priority list, the authority change report and the resource scheduling report, and output a final task planning table.

[0052] In a third aspect, the present invention also provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the employee experience optimization method based on cloud computing and low-code is implemented as described in any one of the above.

[0053] In a fourth aspect, the present invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned employee experience optimization methods based on cloud computing and low-code.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] (1) The process of acquiring employee behavior data, task criticality weight, task safety level, and resource allocation status ensures a comprehensive understanding of employee work status and task environment. Through precise data collection methods, including but not limited to sensor networks and log analysis, employee behavior patterns and various attributes of current tasks can be effectively captured. This step provides a solid data foundation for subsequent sentiment analysis, priority scheduling, authority assessment, and resource scheduling, and improves the data processing efficiency and accuracy of the entire system.

[0056] (2) Performing sentiment analysis based on the employee behavior data to obtain employee sentiment data. This step uses advanced algorithm models (such as deep learning or sentiment computing technology) to extract employee sentiment characteristics from multi-source heterogeneous data, thereby achieving accurate identification of employee sentiment states. The sentiment analysis results can not only reflect the psychological state of employees, but also provide an important reference for subsequent task scheduling, which helps to improve the rationality of task allocation and employee job satisfaction.

[0057] (3) Priority scheduling is performed based on the employee emotional data and the task criticality weights to obtain a priority list. By comprehensively considering the employee emotional data and the task criticality weights, the system can dynamically adjust the execution order of tasks to ensure that critical tasks are assigned to the most suitable employees. This method can not only optimize the timetable for task completion, but also effectively avoid task delays or errors caused by emotional fluctuations, significantly improving the accuracy and efficiency of task scheduling.

[0058] (4) Performing a permission adaptability assessment based on the employee's emotional data and the task security level to obtain a permission change report. Combining the employee's emotional state and the task security level, the system can automatically adjust the employee's access rights to adapt to different work requirements. For example, when an employee is emotionally unstable, his or her access rights to sensitive information are reduced, thereby reducing potential security risks. This permission management strategy based on emotions and security levels enhances the security of the system while ensuring the continuity of work.

[0059] (5) Perform resource scheduling based on the employee emotion data and the resource allocation status to obtain a resource scheduling report. By analyzing employee emotions and their relationship with available resources, the system can reasonably allocate limited resources to ensure that each employee can complete the task in the best condition. For example, when an employee is detected to be in a low mood, the system will arrange more support resources for him, such as additional help or more relaxed time limits. This not only improves the efficiency of resource utilization, but also promotes the work efficiency of employees.

[0060] (6) Generate a task sequence based on the priority list, the authority change report, and the resource scheduling report, and output the final task planning table. The last step integrates the results of all the previous steps and generates a detailed final task planning table through an intelligent algorithm (such as a heuristic algorithm or a constraint satisfaction problem solver). This planning table not only takes into account the urgency of the task, the emotional state of the employee, and the authority level, but also includes the optimal allocation plan for resources. Such comprehensive considerations make task planning more scientific and reasonable, greatly improving work efficiency and the success rate of task execution. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a flowchart of an employee experience optimization method based on cloud computing and low code provided by the first embodiment of the present invention;

[0062] Figure 2 This is a schematic diagram of the structure of an employee experience optimization system based on cloud computing and low code provided by the second embodiment of the present invention. DETAILED DESCRIPTION

[0063] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0064] Reference Figure 1 The first embodiment of the present invention provides an employee experience optimization method based on cloud computing and low code, including the following steps:

[0065] S11, obtain employee behavior data, task criticality weight, task safety level and resource allocation status;

[0066] S12, performing emotion analysis based on the employee behavior data to obtain employee emotion data;

[0067] S13, performing priority scheduling according to the employee emotion data and the task criticality weight to obtain a priority list;

[0068] S14, performing authority adaptability assessment according to the employee emotion data and the task security level, and obtaining an authority change report;

[0069] S15, performing resource scheduling according to the employee emotion data and the resource allocation status to obtain a resource scheduling report;

[0070] S16, generating a task sequence according to the priority list, the authority change report and the resource scheduling report, and outputting a final task planning table.

[0071] In step S11, employee behavior data, task criticality weight, task safety level and resource allocation status are obtained.

[0072] In one implementation, employee behavior data is collected in real time through logging, application program interface (API) integration, or Internet of Things (IoT) sensors on a cloud computing platform, such as recording the frequency of employees' operations on the collaboration platform, task processing time, and error rate, and stored in structured JSON or CSV format for analyzing work patterns and identifying efficiency bottlenecks (e.g., if an employee frequently freezes in a specific process, the interface design can be optimized in a targeted manner); the mission criticality weight is configured by the business department through a low-code platform (e.g., 1-5 points), and stored as a numerical field in the database for dynamic adjustment of task priorities (e.g., setting the customer complaint response weight to 5 points to ensure priority processing); the task security level is graded by the security team based on data sensitivity (e.g., GDPR compliance requirements) and permission policies (e.g., L1-L5), and stored in the form of classification labels or digital codes to ensure that resource allocation complies with security policies (e.g., highly sensitive tasks are only assigned to certified employees); the resource allocation status is captured in real time through the cloud computing platform API, such as server load, license usage, and other data, and stored in a distributed database in JSON format for monitoring resource utilization and triggering automatic capacity expansion (e.g., automatically allocating more bandwidth when insufficient conference system resources are detected). Collaborative analysis of this data can optimize employee workflow, ensure compliance, and improve resource utilization. For example, by combining behavioral data with resource status, additional computing resources can be allocated to critical tasks with high load but low resource support, thereby improving the employee experience.

[0073] In one implementation, employee behavior data includes chat text and voice recordings, which are obtained in the following ways: real-time capture of chat content through the API interface of the company's internal instant messaging tool, or recording of call audio through the recording function of the voice conference system, and after desensitization processing (such as anonymization, removal of sensitive information), it is stored in text form in JSON or CSV format, and the voice data is converted into structured text (such as through voice recognition technology) or encrypted audio files stored in a cloud database (such as AWS S3 or MongoDB). Its purpose is to identify employee emotional states (such as satisfaction, stress or frustration) through natural language processing (NLP) and sentiment analysis algorithms, such as analyzing high-frequency negative words (such as "dissatisfaction" and "unsolvable") in the chat records between customer service staff and customers, or accelerated and hurried speech in the conference voice, thereby triggering managers to intervene (such as providing psychological support or adjusting workload), and ultimately improving employee mental health and team collaboration efficiency.

[0074] In step S12, emotion analysis is performed based on the employee behavior data to obtain employee emotion data.

[0075] In one implementation, text extraction is performed on the voice records in the employee behavior data to obtain voice text; emotion recognition is performed on the voice text and the chat text of the employee behavior data to obtain emotion state labels; an emotion state sequence is constructed based on the emotion state labels at different time periods; standard deviation calculation is performed on the emotion state sequence to obtain an emotion fluctuation amplitude; frequency statistics are performed on the emotion state sequence to obtain an emotion fluctuation frequency; wherein, the employee emotion data includes the emotion fluctuation amplitude and the emotion fluctuation frequency.

[0076] In one implementation, for the text extraction operation, first, the cloud ASR (Automatic Speech Recognition) system preprocesses voice records such as employee meeting recordings and phone conversations, including noise reduction filtering and voice activity detection (VAD) segmentation; then, the pre-trained Wave2Vec 2.0 model is called to convert the voice signal into text, and the CTC decoding algorithm is used to align phonemes and text; finally, non-semantic symbols (such as filler words like "um", "ah", etc.) are filtered out through regular expressions. For example, after processing a certain customer service call, the generated voice text is: "The customer feedbacks that the logistics of order number HD20250316A is delayed. The warehouse has been coordinated to give priority to processing, and it is expected to be delivered within 48 hours."

[0077] In one implementation, emotion recognition uses a multi-task learning model based on the RoBERTa-Large architecture, which is jointly trained on the GoEmotions dataset (containing 28 fine-grained emotion labels) and a self-labeled workplace conversation corpus. The model receives the concatenated voice text and chat records (such as enterprise WeChat conversations) and outputs a composite emotion label containing intensity values. For example, the text of a certain code review meeting "The exception handling mechanism of this module needs to be refactored" is labeled as "Professional advice (intensity 0.7) + Slight anxiety (intensity 0.4)".

[0078] In one implementation, the emotion state sequence is constructed with a 15-minute time window to slice the continuous working period. Within each time window, the emotion labels of multiple texts are integrated through a weighted voting mechanism: the emotion type with the highest intensity value is used as the dominant emotion, and if there is a tie, the multi-label state is retained. For example, the work sequence of a certain R & D personnel in the morning shows a fluctuation trajectory of [Concentration → Confusion → Concentration → Anxiety], and 3 code error discussion records are detected during the "Confusion" period.

[0079] In one implementation, the standard deviation is calculated by counting the degree of dispersion of the dominant emotions in each time period in the quantitative space. The specific process is: first map the emotion label to the valence-arousal two-dimensional coordinate (such as "pleasure" corresponds to +0.8 valence, +0.6 arousal), calculate the square mean of the Euclidean distance between all coordinate points and the mean point, and then take the square root. Frequency statistics record the number of emotion type switches per unit time. For example, a customer service specialist has 23 changes in emotional state during an 8-hour work period. In actual applications, when the amplitude of emotional fluctuations exceeds the threshold of 0.45 (standard deviation unit), a stress warning is triggered, and high-frequency fluctuations (such as more than 4 state switches per hour) indicate the need to adjust the task priority. For example, in a project management case, the on-time delivery rate of high-frequency fluctuation employees is 18.7% lower than that of the stable group.

[0080] In step S13, priority scheduling is performed according to the employee emotion data and the task criticality weight to obtain a priority list.

[0081] In one implementation, the to-do tasks corresponding to the task criticality weights are sorted in descending order according to the task criticality weights to generate an initial task sequence;

[0082] When the emotion fluctuation amplitude in the employee emotion data is greater than a preset amplitude threshold, all long-term planning tasks in the initial task sequence are removed to obtain a short-term task sequence;

[0083] Balance adjustments are made according to the short-term task sequence to obtain a priority list.

[0084] It is worth noting that in the initial task sequence generation stage, the improved TOPSIS multi-criteria decision-making algorithm is used: first, the task criticality weight (0-1 scale) and the task complexity coefficient (derived based on historical completion time regression) are linearly weighted, and the formula is: priority score = 0.7 × criticality weight + 0.3 × (1-complexity coefficient). For example, the pending tasks of the risk control department include: large transaction review (weight 0.95, complexity 0.2), anti-money laundering model training (weight 0.8, complexity 0.6), and department weekly report preparation (weight 0.3, complexity 0.1). After calculation, the scores are 0.905, 0.74, and 0.27 respectively. The task sequence automatically generated by the system puts large transaction review in the first place to ensure that the approval can be completed within 15 minutes.

[0085] It is worth noting that when the emotional fluctuation amplitude exceeds the dynamic threshold (calculated once every hour, taking 1.5 times the standard deviation of the average value of the same position in the last 7 days), the system starts the short-term task filtering engine. The engine identifies task attributes through semantic analysis. For example, when the emotional fluctuation amplitude of a game company's operation and maintenance engineer reaches 0.53 (threshold 0.5), the "disaster recovery architecture upgrade plan (3 months)" in his task queue will be replaced by "real-time server load monitoring".

[0086] It is worth noting that the balance adjustment adopts a mixed integer programming model based on constraint satisfaction, and introduces a workload balance factor while maintaining the criticality of the task. For example, the task sequence optimization of a certain R&D team: the original short-term sequence includes 5 high-intensity code review tasks (average time 90 minutes) and 2 requirements review tasks (120 minutes). After the system detects that the frequency of employee emotional fluctuations reaches 3.2 times per hour, it automatically inserts 2 low cognitive load document writing tasks (30 minutes each), forming an alternating sequence of [review → document → review → review]. Actual operation data shows that the adjusted task sequence reduces the code review error rate from 15% to 7%, while the daily task completion volume increases by 22%. The core of the balance adjustment is to insert low cognitive load buffer tasks between critical tasks, thereby reducing employee fatigue by dispersing the workload, and ultimately improving the overall task completion efficiency.

[0087] It is worth noting that the workload balance factor dynamically adjusts the order of tasks by comprehensively evaluating the importance of tasks and the actual bearing capacity of employees. Its calculation logic is: first, according to the priority of the task (such as code review, which is a key task with a high priority), the execution order is determined to be unshakable, and then based on the current state of the employee (such as the frequency of emotional fluctuations, fatigue) and the cognitive load of the task (such as high-intensity tasks require high concentration and take a long time), the "load pressure value" of the employee under the current task sequence is calculated. For example, the load value of high-intensity tasks is set to 1.0, and the low-intensity tasks are set to 0.3, and the safety upper limit is set in combination with the emotional fluctuation threshold (such as 1.5 fluctuations are allowed per hour). When the system detects that the load pressure value of the original task sequence (such as 5 consecutive high-intensity code reviews) exceeds the threshold (such as emotional fluctuations of 3.2 times / hour), it will automatically insert low-load tasks (such as document writing) between high-load tasks, and by reducing the average load value per unit time (such as reducing the load per hour from 1.33 to 0.65) and reducing emotional fluctuations (to 1.2 times / hour), it will ultimately ensure the priority of key tasks while achieving a reasonable distribution of work intensity, thereby improving efficiency and quality.

[0088] In one implementation, the workload balancing factor is calculated as an average load value per unit time.

[0089] In step S14, an authority adaptability assessment is performed based on the employee emotion data and the task security level to obtain an authority change report.

[0090] In one implementation, historical permission records are obtained; permission thresholds are calculated based on the historical permission records and the task security level to obtain a permission threshold; when the frequency of emotional fluctuations in the employee emotional data is greater than the permission threshold, the task permission corresponding to the task security level is adjusted to inaccessible; when the frequency of emotional fluctuations is less than the permission threshold, the task permission is adjusted to accessible; wherein the permission change report includes the task permission and the corresponding change record.

[0091] In one implementation, the permission adjustment process implements a double verification mechanism: when it is detected that the frequency of fluctuations exceeds the standard, the system first freezes sensitive permissions and synchronously updates the AD domain control through the LDAP protocol. For example, in the operation log audit task (confidential level) of a nuclear power control center, the operator Zhang had 5.2 mood swings (threshold 4.1 times) during the period of 10:00-11:00 in the morning, and his SCADA system level 3 operation permission was temporarily downgraded to read-only mode. The log shows that the core temperature calibration task that Zhang was responsible for during this period was automatically transferred to the standby engineer. The system recorded 9 illegal access attempts and triggered a Kerberos authentication alarm.

[0092] In one implementation, the permission change report uses blockchain evidence storage technology and includes a three-dimensional permission matrix: 1) Data dimension: For example, the access rights to the compound database of a pharmaceutical company's R&D center are adjusted from "full control" to "query only"; 2) Operation dimension: When the fluctuation frequency of the financial department's fund transfer authority reaches the threshold, the single limit is reduced from 5 million to 500,000; 3) Time dimension: The operation and maintenance window of cloud computing engineers is adjusted from 7×24 hours to 09:00-17:00 on working days during abnormal mood periods. A typical case shows that the high-precision map update authority of an autonomous driving company was disabled due to abnormal mood fluctuations of engineers (frequency 6.8 times / hour), resulting in the disabling of 3 sensitive API interfaces. The system automatically generates an audit report containing 12 fine-grained permission changes and pushes it to the compliance department.

[0093] In one implementation, the calculating of the permission threshold according to the historical permission record and the task security level to obtain the permission threshold includes:

[0094] The permission threshold is calculated using the following formula:

[0095]

[0096] in, express The permission threshold at the moment, Indicates the current moment, Represents the historical permission baseline, represents the historical authority coefficient, express The real-time entropy weight at the moment, Indicates the security level is The safety level adjustment factor is Indicates the task safety level, represents the coefficient of deviation, represents the natural base, Indicates the absolute deviation of the current authority value from the historical mean.

[0097] It is worth noting that the calculation of the permission threshold dynamically adjusts the permission level by integrating historical data, real-time dynamics and security constraints. The historical permission baseline is the average value of employee permissions over a period of time in the past. For example, the average permission value in the past month is 50, which serves as a basic reference; the historical permission coefficient is a weight between 0 and 1. For example, it is set to 0.7, which means that historical data accounts for 70% of the weight to maintain the stability of permissions; the real-time entropy weight reflects the complexity or uncertainty of the current task. For example, the real-time entropy weight is 60, which means that the current task has a high degree of uncertainty; the security level adjustment factor is dynamically adjusted according to the security level of the task. For example, the adjustment factor of a high-security level task (such as processing sensitive data) is set to 0.8, which means that the permission threshold needs to be reduced by 20% to enhance security; the deviation coefficient is a positive number, for example, it is set to 0.005, which is used to control the intensity of the impact of permission deviation on the result; the absolute deviation of the current permission value from the historical mean is the difference between the current permission value and the historical mean. For example, if the current permission value is 60 and the historical average is 50, the deviation is 10. Calculated by the formula: First, calculate the historical and real-time weighted average of 53, then multiply it by the safety adjustment factor to get 42.4, and finally pass the exponential term ( The final permission threshold is about 25.7. This process ensures that permissions are stable based on historical behavior and can be flexibly adjusted according to real-time risks and security requirements. For example, when a task involves highly sensitive data and employee permissions suddenly increase, the system will automatically tighten permissions to avoid security risks. This approach not only helps maintain the security of the system, but also adapts to changes in permission requirements in different business scenarios, improving overall work efficiency and employee experience.

[0098] In step S15, resource scheduling is performed according to the employee emotion data and the resource allocation status to obtain a resource scheduling report.

[0099] In one implementation, idle calculations are performed based on the resource allocation status to obtain idle computing resources; when the frequency of emotion fluctuations in the employee emotion data is greater than a preset frequency threshold, the idle computing resources are allocated to the employee; when the emotion fluctuation data is less than a preset frequency threshold, resource recovery calculations are performed based on the employee emotion data to obtain a recovery resource ratio; based on the recovery resource ratio, the computing resources allocated to the employee are recovered and the resource allocation status is updated; wherein the resource scheduling report includes resource scheduling records and resource allocation status.

[0100] In one implementation, when the frequency of employee emotional fluctuations is lower than a preset threshold, the system calculates the resource recovery ratio through the following steps: First, the total amount of idle computing resources is counted according to the current resource allocation status, and the basic recovery ratio is determined in combination with employee emotional data (such as the duration of the fluctuation frequency below the threshold, the improvement of task completion efficiency) (for example, the basic ratio increases by 5% for every 0.1 times / hour that the fluctuation frequency is lower than the threshold). Secondly, the priority of the employee's current task and the actual resource utilization rate are analyzed. If the resource occupancy rate of the high-priority task is lower than 80% or the idle time of the low-priority task exceeds the preset duration (such as 2 hours), the recovery ratio is further increased (for example, 10% is added). Finally, the basic ratio is added to the dynamically adjusted value to obtain the total recovery ratio (such as basic ratio 15% + dynamic adjustment 10% = 25%), and idle resources are recovered according to this ratio (such as 25 units of resources are recovered if 100 units of resources were originally allocated), while ensuring that the minimum resource requirements of key tasks are not affected, and the recovery operation is recorded in the resource scheduling report to update the allocation status.

[0101] In one implementation, during the idle resource calculation phase, the system uses dynamic resource map construction technology: the container orchestration platform collects CPU / GPU utilization, memory occupancy and other indicators in real time, and combines Kubernetes' Horizontal PodAutoscaler algorithm to make idle judgments. For example, in the production environment of a cloud service provider, when it is detected that the GPU utilization of an AI training cluster is lower than 35% for 30 consecutive minutes (the preset threshold is 40%), the system marks 32 NVIDIA A100 nodes as schedulable resources. A specific case shows that the model training task of an autonomous driving company therefore obtained an additional 48TFLOPS computing power support, which shortened the completion time of night batch jobs from 9.2 hours to 5.5 hours.

[0102] In one implementation, when emotional fluctuations trigger resource allocation, the system implements an intelligent bandwidth adjustment mechanism: when the emotional monitoring bracelet detects that the standard deviation of a customer service specialist's heart rate variability (HRV) exceeds 72ms (corresponding to a frequency threshold of 3.8 times / hour), it automatically allocates a backup call channel to its session system through the OpenStack API. The practice of a bank's credit card center shows that during the peak period of emotional fluctuations in the afternoon (14:00-16:00), this mechanism increases concurrent processing capacity by 40%, and the median customer waiting time is reduced from 127 seconds to 79 seconds. The system log shows that during a major promotion, the group of employees with abnormal emotions (fluctuation frequency 4.2±0.3 times) was allocated an additional 128 vCPU cores, ensuring a 98.7% SLA compliance rate.

[0103] In one implementation, the resource recovery process adopts a progressive backoff strategy: when the emotional stability index (ESI) of the development team remains above 85 points for 2 consecutive hours (frequency threshold 2.1 times), the system starts a resource recovery algorithm based on a sliding time window. An example of a multinational software company shows that its CI / CD pipeline automatically recovers 20%-35% of Jenkins executor resources in the early morning hours every day, and dynamically adjusts the recovery ratio through a reinforcement learning model. Specific data shows that the compilation resource pool of a Java microservice team recovered 24 container instances during the emotional stability period (fluctuation frequency 1.8 times), and these resources were reallocated to emergency security patch tasks, which increased the vulnerability repair response speed by 63%. The resource scheduling report uses a time series database to store change records, and generates detailed logs every 5 minutes, including 12 dimensions such as resource type, operation type (allocation / recycling), and impact range.

[0104] It is worth noting that the resource recovery algorithm monitors the emotional stability index (ESI) of the development team and the frequency of resource usage fluctuations in real time. After the trigger conditions are met (ESI is ≥85 minutes for 2 consecutive hours and the frequency threshold is ≥2.1 times / hour), it uses the sliding time window technology to continuously analyze the recent resource utilization data, and combines the reinforcement learning model to dynamically calculate the optimal recovery ratio (such as 20%-35% in the example). During specific execution, the system generates recovery instructions in proportion to the total amount of the current resource pool (for example, the recovery decision of 24 container instances of a team), giving priority to idle resources occupied by non-critical tasks, and recording 12-dimensional log data such as resource type and operation type every 5 minutes through the time series database, providing a feedback loop for subsequent model optimization, and finally achieving a flexible balance between resource recovery and emergency task allocation.

[0105] In step S16, a task sequence is generated according to the priority list, the authority change report and the resource scheduling report, and a final task planning table is output.

[0106] In one implementation, permissions are matched based on the priority list and the permission change report to obtain a permission adaptation task sequence; sorting optimization is performed based on the resource scheduling report and the permission adaptation task sequence to generate a preliminary execution table; historical execution deviation data is obtained; fault tolerance correction is performed based on the historical execution deviation data, and a final task planning table is output.

[0107] In one implementation, during the permission matching phase, the system uses attribute-based encryption (ABE) technology to dynamically parse the three-dimensional permission matrix (operation object, permission type, time validity) in the permission change report, and matches the security attributes (such as data sensitivity, operation risk level) in the task priority list with the current permission status through the security level mapping mechanism. The specific process includes: based on the task permission attributes (such as operation object, permission type) and their change records (upgrade / downgrade status) in the permission change report, the system compares the current user permission with the security level required by the task (such as data sensitivity, operation risk level) in real time. When a permission mismatch is detected (such as the operator lacks L4 permission), the task execution is immediately interrupted and the permission adaptation mechanism is automatically triggered - through the preset rule library, a backup execution subject that meets the conditions is selected (such as switching to a compliance audit specialist within 0.5 seconds) or a permission upgrade application is initiated, and the permission verification result, adaptation operation type (redirection / adjustment) and operation timestamp are written into the blockchain log to form a complete permission adaptation task sequence. For example, before executing a bank's anti-money laundering monitoring task (security level L4), it was detected that the operator's "cross-border transaction data export" authority was downgraded. The system rerouted the task to a compliance audit specialist with L4 authority within 0.5 seconds and generated a record of authority transfer on the blockchain to ensure that the operation was traceable.

[0108] In one implementation, the sorting optimization uses a mixed integer programming (MIP) algorithm, combined with real-time resource status data (CPU / GPU utilization, memory occupancy, etc.) in the resource scheduling report to construct the objective function and constraints. The specific process is divided into four steps: ① Establish a mathematical model that includes task execution time, resource demand coefficient, and deadline penalty weight; ② Solve the optimal task sorting through the branch and bound method; ③ Apply the time window constraint satisfaction algorithm (TW-CSP) to coordinate cross-regional resources; ④ Generate a preliminary execution table containing task ID, resource allocation plan, and expected completion time. The core function of this table is to achieve dynamic resource balance. For example, in a cloud data center log audit task, when real-time monitoring shows that the GPU resource pool utilization reaches the 85% threshold, the system automatically compresses the Tensor Core occupancy rate of non-critical tasks from 32% to 18%, giving priority to the 128GB video memory requirement of the payment gateway, and increasing the SLA compliance rate of core business to 99.98%.

[0109] In one implementation, fault tolerance correction builds a prediction model by analyzing the historical execution deviation database (including task delay records, resource conflict events, and permission exception logs for the past 180 days). The system uses LSTM neural networks to identify 8 typical deviation patterns. For example, in the securities trading settlement scenario, when the composite deviation feature of "permission change-resource over-allocation" is detected (historical probability of occurrence 23.7%), a three-level fault tolerance mechanism is triggered: first, the task deadline is recalculated based on the sliding time window algorithm (such as the 20 million transaction settlement tasks originally scheduled to be completed at 03:00 are flexibly extended to 03:15), and then the backup task node is dynamically inserted (a redundant settlement engine deployed in a remote data center), and finally a final planning table containing 9 fault tolerance clauses is generated. After applying this mechanism, an automobile manufacturer successfully reduced the handover period deviation rate of material scheduling tasks from 37% to 8.5%, and achieved zero-delay production by pre-allocating 5 spare AGV transport vehicles.

[0110] In summary, the present invention discloses an employee experience optimization method based on cloud computing and low-code, which aims to optimize the employee work experience by analyzing employee emotions, scheduling task priorities, task permissions, and computing resources.

[0111] Reference Figure 2 The second embodiment of the present invention provides an employee experience optimization system based on cloud computing and low code, including:

[0112] Data acquisition module, used to obtain employee behavior data, task criticality weight, task safety level and resource allocation status;

[0113] A sentiment analysis module, used to perform sentiment analysis based on the employee behavior data to obtain employee sentiment data;

[0114] A weight scheduling module, used to perform priority scheduling according to the employee emotion data and the task criticality weight to obtain a priority list;

[0115] The permission change module is used to evaluate the suitability of permissions based on the employee emotion data and the task security level to obtain a permission change report;

[0116] A resource scheduling module, used to perform resource scheduling according to the employee emotion data and the resource allocation status, and obtain a resource scheduling report;

[0117] The task planning module is used to generate a task sequence according to the priority list, the authority change report and the resource scheduling report, and output a final task planning table.

[0118] It should be noted that an employee experience optimization system based on cloud computing and low code provided in an embodiment of the present invention is used to execute all the process steps of an employee experience optimization method based on cloud computing and low code in the above-mentioned embodiment. The working principles and beneficial effects of the two correspond one to one, and therefore will not be repeated here.

[0119] An embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a data acquisition program. When the processor executes the computer program, the steps in the above-mentioned embodiments of the employee experience optimization method based on cloud computing and low code are implemented, such as Figure 1 Alternatively, when the processor executes the computer program, the functions of the modules / units in the above-mentioned device embodiments are realized, such as the data acquisition module.

[0120] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the electronic device.

[0121] The electronic device may be a computing device such as a desktop computer, a notebook, a PDA, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. The electronic device may include more or fewer components than the above components, or may combine certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0122] The processor may be a central processing unit (CPU), 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. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, and uses various interfaces and lines to connect various parts of the entire electronic device.

[0123] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the electronic device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0124] Wherein, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0125] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0126] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for optimizing employee experience based on cloud computing and low code, characterized in that: include: Obtain employee behavior data, task criticality weights, task safety levels, and resource allocation status; Performing sentiment analysis based on the employee behavior data to obtain employee sentiment data; Performing priority scheduling according to the employee emotion data and the task criticality weights to obtain a priority list; Perform authority adaptability assessment based on the employee emotion data and the task security level to obtain an authority change report; Perform resource scheduling according to the employee emotion data and the resource allocation status to obtain a resource scheduling report; A task sequence is generated according to the priority list, the authority change report and the resource scheduling report, and a final task planning table is output.

2. The employee experience optimization method based on cloud computing and low code according to claim 1 is characterized in that: The performing of emotion analysis based on the employee behavior data to obtain employee emotion data includes: Extracting text from the voice records in the employee behavior data to obtain voice text; Performing emotion recognition based on the voice text and the chat text of the employee behavior data to obtain an emotion state label; Constructing an emotional state sequence according to the emotional state labels of different time periods; Calculating the standard deviation of the emotional state sequence to obtain the emotional fluctuation amplitude; Performing frequency statistics on the emotional state sequence to obtain the frequency of emotional fluctuations; The employee emotion data includes the emotion fluctuation amplitude and the emotion fluctuation frequency.

3. The employee experience optimization method based on cloud computing and low code according to claim 1 is characterized in that: The priority scheduling is performed according to the employee emotion data and the task criticality weight to obtain a priority list, including: According to the task criticality weights, the to-do tasks corresponding to the task criticality weights are sorted in descending order to generate an initial task sequence; When the emotion fluctuation amplitude in the employee emotion data is greater than a preset amplitude threshold, all long-term planning tasks in the initial task sequence are removed to obtain a short-term task sequence; Balance adjustments are made according to the short-term task sequence to obtain a priority list.

4. The employee experience optimization method based on cloud computing and low code according to claim 1 is characterized in that: The authority adaptability assessment is performed according to the employee emotion data and the task security level to obtain an authority change report, including: Get historical permission records; Calculate the authority threshold according to the historical authority record and the task security level to obtain the authority threshold; When the frequency of emotional fluctuations in the employee's emotional data is greater than the authority threshold, the task authority corresponding to the task security level is adjusted to be inaccessible; When the frequency of the emotion fluctuation is less than the permission threshold, adjusting the task permission to be accessible; The permission change report includes the task permission and the corresponding change record.

5. The employee experience optimization method based on cloud computing and low code according to claim 4 is characterized in that: The calculating the authority threshold according to the historical authority record and the task security level to obtain the authority threshold includes: The permission threshold is calculated using the following formula: ； in, express The permission threshold at the moment, Indicates the current moment, Represents the historical permission baseline, represents the historical authority coefficient, express The real-time entropy weight at the moment, Indicates the security level is The safety level adjustment factor is Indicates the task safety level, represents the coefficient of deviation, represents the natural base, Indicates the absolute deviation of the current authority value from the historical mean.

6. The employee experience optimization method based on cloud computing and low code according to claim 1 is characterized in that: The resource scheduling is performed according to the employee emotion data and the resource allocation status to obtain a resource scheduling report, including: Performing idle computing according to the resource allocation state to obtain idle computing resources; When the frequency of emotion fluctuations in the employee emotion data is greater than a preset frequency threshold, allocating the idle computing resources to the employee; When the emotion fluctuation data is less than a preset frequency threshold, a resource recovery calculation is performed based on the employee emotion data to obtain a resource recovery ratio; Reclaiming computing resources allocated to employees according to the resource recovery ratio, and updating the resource allocation status; The resource scheduling report includes resource scheduling records and resource allocation status.

7. The employee experience optimization method based on cloud computing and low code according to claim 1 is characterized in that: The step of generating a task sequence according to the priority list, the authority change report and the resource scheduling report and outputting a final task planning table includes: Performing permission matching according to the priority list and the permission change report to obtain a permission adaptation task sequence; Perform sorting optimization according to the resource scheduling report and the permission adaptation task sequence to generate a preliminary execution table; Get historical execution deviation data; Fault tolerance correction is performed based on the historical execution deviation data, and a final task planning table is output.

8. An employee experience optimization system based on cloud computing and low code, characterized in that: include: Data acquisition module, used to obtain employee behavior data, task criticality weight, task safety level and resource allocation status; A sentiment analysis module, used to perform sentiment analysis based on the employee behavior data to obtain employee sentiment data; A weight scheduling module, used to perform priority scheduling according to the employee emotion data and the task criticality weight to obtain a priority list; The permission change module is used to evaluate the suitability of permissions based on the employee emotion data and the task security level to obtain a permission change report; A resource scheduling module, used to perform resource scheduling according to the employee emotion data and the resource allocation status, and obtain a resource scheduling report; The task planning module is used to generate a task sequence according to the priority list, the authority change report and the resource scheduling report, and output a final task planning table.

9. An electronic device, characterized in that: It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the cloud computing and low-code based employee experience optimization method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein, when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the cloud computing and low-code based employee experience optimization method as described in any one of claims 1 to 7.

Images