Multi-module network quality self-inspection method based on perception marketing transformation index

Abstract

The invention provides a multi-module network quality self-inspection method based on a perception marketing transformation index, and relates to the technical field of network quality detection.The method comprises the steps that various kinds of basic data in network operation are acquired through data collection, and the basic data comprise a network perception degradation index, a terminal perception degradation index, a perception marketing portrait index and the like; processing and analyzing the basic data through a perceptual marketing transformation index calculation model to obtain a perceptual marketing transformation index; determining a network quality grade according to the index and a preset corresponding relation, and generating an evaluation report; and when the quality is poor, the root of the problem is accurately positioned, and finally, the optimization suggestion generation module gives targeted optimization suggestions in combination with the diagnosis result. The network quality can be automatically and comprehensively self-inspected, a powerful basis is provided for network optimization, and the network quality and the marketing conversion effect are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of network quality detection, and in particular to a multi-module self-inspection network quality method based on a perceived marketing conversion index. Background Art

[0002] With the rapid development of the internet industry, network quality has a crucial impact on user experience and marketing conversion effectiveness. Traditional network quality assessment methods often focus on technical network performance indicators alone, such as bandwidth, latency, and packet loss rate, to determine network quality. However, this approach has significant limitations, ignoring the relationship between network terminals, network equipment, and other factors, as well as their actual impact on marketing conversion.

[0003] From the perspective of network terminals, device characteristics can reflect the user's true needs and experience in the network environment. For example, traditional assessment methods fail to fully consider the network quality information contained in these terminal data, such as excessive terminal CPU utilization, excessive terminal memory usage, long periods of terminal rebooting, and excessive terminal temperature.

[0004] Furthermore, in the fiercely competitive market, the assessment of perceived marketing profiles is crucial for network quality. As a description of the user's primary persona, the perceived marketing profile's index directly impacts marketing conversion effectiveness. Previous marketing conversion evaluations have mostly focused on marketing strategies and advertising, rarely incorporating the perceived marketing profile index as a key variable into comprehensive considerations. This has hindered accurate assessment of the role of network quality in marketing conversion.

[0005] At the same time, existing network quality monitoring methods are mostly single-dimensional and passive, lacking a network quality assessment mechanism that can proactively, comprehensively, and dynamically integrate multiple factors for self-inspection and provide specific optimization guidance. This makes it difficult for network operators to accurately identify the root cause of network quality issues, forcing them to rely on experience for troubleshooting and optimization. This is not only inefficient but also prone to missing key issues, making it difficult to effectively improve network quality and marketing conversion results. Summary of the Invention

[0006] To solve the above problems, the present invention provides a multi-dimensional active network quality monitoring method, which can comprehensively and dynamically perform self-inspection on multiple factors and provide specific optimization guidance methods.

[0007] To achieve the above object, the technical solution adopted by the present invention is: In a first aspect of the present invention, a multi-module self-checking network quality method based on a perceived marketing conversion index is provided, comprising the following steps: Collect basic data during network operations, including the network perception degradation index, terminal perception degradation index, and perception marketing portrait index; Based on the preset perception marketing conversion index model, the collected basic data is analyzed and processed to calculate the perception marketing conversion index; Evaluate network quality based on the calculated perceived marketing conversion index; When the network quality assessment result is lower than the preset threshold, the problem is analyzed through problem diagnosis methods and the fault location function is used to locate the fault; Based on the results of problem diagnosis and location, combined with the preset optimization strategy library, optimization suggestions for current network quality issues are generated.

[0008] Preferably, the network perception degradation index is obtained by comprehensive judgment of the QoE perception score, the number of optical modem restarts, the Lan port negotiation rate, the intranet device negotiation rate, the number of PON and WAN interruptions and the received optical power.

[0009] Preferably, the terminal perception degradation index is obtained by comprehensively judging the terminal CPU usage, the terminal memory resource occupancy, the terminal restart interval and the terminal temperature.

[0010] Preferably, the perceptual marketing portrait index is obtained by comprehensively judging the number of devices connected to the intranet, the number of smart home devices in the intranet, the number of high-value mobile devices, the number of self-purchased 100M old routers, the number of home routers, the number of home Wi-Fi amplifiers and the number of home Linux servers.

[0011] Preferably, the perception marketing conversion index is obtained by weighted combination of the network perception degradation index, the terminal perception degradation index and the perception marketing portrait index, as follows: in, represents the marketing conversion index, represents the network perception degradation index, represents the terminal perception degradation index, represents the perceived marketing portrait index, 、 and represents the weight, where and = =0.2, .

[0012] Preferably, the perception marketing conversion index model is trained using a machine learning algorithm, the training data includes historical network operation data and corresponding marketing conversion result data, and the model parameters are optimized through continuous iteration.

[0013] Preferably, the network quality is evaluated by: in, Indicates the current network quality score. Indicates the perceived marketing portrait index value calculated in the current cycle. Represents a user Current network transmission rate, Indicates the mapping quality score between the perceived marketing profile index value and the network transmission rate, represents the influence coefficient, represents the error delay function.

[0014] In a second aspect of the present invention, a multi-module self-checking network quality system based on a perceived marketing conversion index is provided, comprising: The data collection module is used to collect basic data during network operation, including the network perception degradation index, terminal perception degradation index, and perception marketing portrait index; The perceptual marketing conversion index model construction module is used to analyze and process the collected basic data based on the construction of the perceptual marketing conversion index model to calculate the perceptual marketing conversion index; A network quality evaluation module is used to evaluate the network quality based on the calculated perceived marketing conversion index; The problem diagnosis and location module is used to analyze the problem through problem diagnosis methods and locate the fault location through fault location functions when the network quality assessment result is lower than the preset threshold; The optimization suggestion generation module is used to generate optimization suggestions for current network quality issues based on the results of problem diagnosis and location, combined with the preset optimization strategy library.

[0015] The beneficial effects of the present invention are: Accurately assessing network quality: By comprehensively considering multiple dimensions, including the Network Perception Degradation Index, the Terminal Perception Degradation Index, and the Perception Marketing Profile Index, and utilizing the Perception Marketing Conversion Index as a quantitative metric, this method more accurately reflects the impact of actual network operational quality on marketing conversion. Compared to traditional assessments based solely on network technical parameters, this method provides a more comprehensive view of network quality, avoids assessment bias caused by ignoring key factors, and provides more targeted guidance for network optimization.

[0016] Multi-module collaboration improves efficiency: Each module has a clear division of labor and close collaboration. The data collection module collects a wide range of basic data to ensure comprehensiveness and accuracy. The perception marketing conversion index calculation module leverages scientific model algorithms to deeply explore the value of this data. The network quality assessment module quickly provides quantitative assessment results based on pre-set standards. The problem diagnosis and location module accurately identifies the source of the problem, eliminating the need for large-scale manual investigation. The optimization suggestion generation module directly outputs practical improvement measures based on the diagnosis results. The entire process is automated and systematized, significantly improving the efficiency of network quality self-inspection and optimization work, saving manpower and material costs.

[0017] Enhanced Marketing Conversion: Focusing on marketing conversion as the core goal, we closely link network quality with marketing activities, promoting increased marketing conversion by improving network quality. When the network better meets user needs and provides a higher-quality experience, user participation and acceptance of marketing activities will increase accordingly, ultimately increasing the company's marketing effectiveness and enhancing its competitiveness in the market.

[0018] Real-time Monitoring and Dynamic Optimization: The frequency and scope of data collection can be flexibly adjusted based on actual needs, enabling real-time monitoring of network quality. Once an anomaly or declining trend in network quality is detected, a self-check process is triggered to quickly locate the issue and provide optimization suggestions, ensuring the network remains optimal and adapts to the ever-changing network environment and sustainable business development.

[0019] Wide Applicability and Scalability: This method and system are applicable to a variety of network application scenarios and personal and enterprise network architectures of varying sizes. Whether it's an e-commerce website, an online service platform, or an advertising marketing network, simple adjustments to the data collection scope and model parameters based on specific business characteristics are all required for operational implementation. Furthermore, as business grows and technology advances, each module can be further expanded and improved, such as by incorporating more advanced data analysis algorithms into the Perception Marketing Conversion Index calculation module, continuously optimizing and enhancing the system's performance and functionality. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a flow chart of a multi-module self-inspection network quality method based on the perception marketing conversion index of the present invention.

[0021] Figure 2 This is a framework diagram of a multi-module self-inspection network quality system based on the perception marketing conversion index of the present invention.

[0022] Figure 3 This is a diagram showing the network perception degradation index of a preferred embodiment of the present invention.

[0023] Figure 4This is a diagram showing the terminal perception degradation index of a preferred embodiment of the present invention.

[0024] Figure 5 This is a display diagram of the perceptual marketing portrait index of a better embodiment of the present invention. DETAILED DESCRIPTION

[0025] This example uses the network operations of a large e-commerce platform as an example. This platform faces massive user traffic and frequent marketing activities. The quality of its network is directly related to user experience and marketing conversion effectiveness. Therefore, the platform decided to adopt the multi-module self-inspection network quality method and system based on the perceived marketing conversion index proposed in this patent to improve network operation quality.

[0026] See also Figure 1 As shown, in a first aspect of the present invention, a multi-module self-checking network quality method based on the perceived marketing conversion index is provided, comprising the following steps: Collect basic data during network operations, including the network perception degradation index, terminal perception degradation index, and perception marketing portrait index; Basic data is collected through the following steps: User behavior data collection: We collect user behavior data through tracking technology at various user access points, such as the e-commerce platform's mobile app and website. This data includes details such as the length of time users browse product pages, how often they click on products, how many times they add products to their shopping carts, and their purchase conversion behavior. Furthermore, sensors on users' mobile devices monitor user behavior habits, such as touch force and scrolling speed, to further enrich this user behavior data.

[0027] Network performance data collection: Deploy network performance monitoring tools on the e-commerce platform's servers and network switches to collect real-time key network performance indicators such as network bandwidth usage, server response time, page loading time, data transmission packet loss rate, terminal CPU usage, terminal memory resource utilization, terminal restart interval and terminal temperature, number of optical modem restarts, Lan port negotiation rate, intranet device negotiation rate, PON and WAN interruption times, and received optical power.

[0028] Marketing campaign data collection: Obtain relevant data on marketing campaigns from the marketing campaign management system of the e-commerce platform, such as advertising channels, ad exposure, click-through rate, number of participants in promotional activities, issuance and use of coupons, number of offline devices, number of smart home devices on the intranet, number of high-value mobile devices, number of self-purchased 100M old routers, number of home routers, number of home Wi-Fi amplifiers, and number of home Linux servers.

[0029] For user behavior data, we cross-check the data collected from the mobile app with the data from the web page. For example, if a user browses a product page on one device and then also browses the same product page on another device, but the browsing time is significantly abnormal (e.g., 10 minutes on the mobile device but only 1 second on the web page), we determine that the abnormal data may be erroneous and will be further verified or eliminated.

[0030] Comparison of authoritative network monitoring tools: For network performance data, we simultaneously use multiple authoritative network monitoring tools (e.g., Network Performance Monitoring System A and Network Performance Monitoring System B) to collect metrics like network bandwidth and latency. We then compare the data collected by the different tools for the same metric. If the difference is within a reasonable margin of error (e.g., ±5%), the data is considered accurate. If the difference is significant, we investigate whether it's a malfunction in one of the monitoring tools or a unique aspect of the network environment. We then address the issue and collect data again.

[0031] Among them, the network perception degradation index, terminal perception degradation index and perception marketing portrait index are as follows: Figure 3-5 As shown, the collected basic data is analyzed and processed to calculate the perceived marketing conversion index.

[0032] Based on the preset perception marketing conversion index model, the collected basic data is analyzed and processed to calculate the perception marketing conversion index; Based on the pre-established perceptual marketing conversion index model, this model comprehensively considers user behavior characteristic indicators (such as user browsing time, click frequency, purchase conversion rate, etc.), network performance indicators (such as page loading speed, server response time, etc.) and marketing activity delivery indicators (such as ad clicks, promotion activity participation, etc.). It is obtained by training historical data through machine learning algorithms and can accurately reflect the relationship between marketing conversion effects and various influencing factors under network conditions.

[0033] The Perceptual Marketing Conversion Index model is trained using a machine learning algorithm. The training data includes historical network operation data and corresponding marketing conversion result data. The model parameters are optimized through continuous iteration.

[0034] The network perception degradation index is obtained through comprehensive judgment of the QoE perception score, the number of optical modem restarts, the Lan port negotiation rate, the intranet device negotiation rate, the number of PON and WAN interruptions, and the received optical power.

[0035] The terminal perception degradation index is obtained through comprehensive judgment of the terminal CPU usage, terminal memory resource usage, terminal restart interval and terminal temperature.

[0036] The perceptual marketing portrait index is obtained through a comprehensive assessment of the number of devices connected to the intranet, the number of smart home devices on the intranet, the number of high-value mobile devices, the number of self-purchased 100M old routers, the number of home routers, the number of home Wi-Fi amplifiers, and the number of home Linux servers.

[0037] The basic data collected by the data collection module is input into the model. After being analyzed and processed by the model, the Figure 3-5 The method shown is used to calculate the perceived marketing conversion index.

[0038] The Perception Marketing Conversion Index is obtained through a weighted combination of the Network Perception Degradation Index, the Terminal Perception Degradation Index, and the Perception Marketing Portrait Index, as follows: in, represents the marketing conversion index, represents the network perception degradation index, represents the terminal perception degradation index, represents the perceived marketing portrait index, 、 and represents the weight, where and = =0.2, .

[0039] For example, in a certain marketing campaign, based on the collected data of few optical modem restarts, terminal CPU usage, terminal memory resource occupancy, terminal restart interval and terminal temperature lower than the average, and a small number of intranet-connected devices and intranet smart home devices, the model calculated a perceived marketing conversion index of 85 (out of 100), indicating that the marketing conversion effect is at a good level under the current network conditions.

[0040] Evaluate network quality based on the calculated perceived marketing conversion index; Based on a pre-defined correlation between the Perceived Marketing Conversion Index and network quality levels (e.g., 90-100 is excellent, 80-89 is good, 70-79 is fair, 60-69 is poor, and below 60 is unsatisfactory), a calculated Perceived Marketing Conversion Index of 85 is classified as "good" and a network quality assessment report is generated. This report details the network quality level, key influencing factors (e.g., page load speed needs improvement, some advertising channels have average conversion performance), and recommended improvements (e.g., optimizing server configuration to increase page load speed, adjusting advertising strategies, etc.).

[0041] Network quality is assessed by: in, Indicates the current network quality score. Indicates the perceived marketing portrait index value calculated in the current cycle. Represents a user Current network transmission rate, Indicates the mapping quality score between the perceived marketing profile index value and the network transmission rate, represents the influence coefficient, represents the error delay function.

[0042] This design introduces an error delay penalty term based on the data matching of network quality detection. It targets the situation where device quality causes network errors in real network scenarios. For example, the functions of WiFi5 and WiFi6 devices are the same but the transmission rates are very different. In addition, by introducing the error delay function term, , which can effectively predict the probability of errors caused by temperature and overload after long-term operation of the equipment.

[0043] represents the error delay function, as the device running time increases, The value will increase, indicating that the device has been running for a long time and has errors caused by temperature and overload.

[0044] For example: Assume that a device performs well in the early stage, but due to the increase of network devices and user access, the terminal CPU usage and terminal memory resource usage are too high, and the terminal restart interval may become longer, resulting in a decrease in network quality. In this case, It will increase over time, thereby dynamically adjusting the network quality score so that the network quality score more accurately reflects the actual performance of the current network quality.

[0045] When the network quality assessment result is lower than the preset threshold, the problem is analyzed through problem diagnosis methods and the fault location function is used to locate the fault; This module automatically activates when the network quality assessment result is "Medium" or lower. By analyzing abnormal fluctuations in basic data indicators and their correlations, it pinpoints the specific links and factors contributing to the decline in network quality. For example, in one assessment, the Perceived Marketing Conversion Index dropped to 70. The Problem Diagnosis and Location Module identified a significant increase in server response time and loading errors on some pages. Further investigation determined that insufficient server storage space and peak-time network bandwidth congestion were the cause.

[0046] Continuously monitor the changing trends of various basic data indicators, and promptly detect abnormal fluctuations by setting reasonable thresholds and fluctuation ranges.

[0047] For example, if the page bounce rate in user behavior data suddenly increases significantly, or the server response time in network performance data is significantly beyond the normal range, these abnormal fluctuations can be used as clues to potential problems for in-depth analysis.

[0048] The data is then cross-analyzed from multiple dimensions, exploring potential correlations between the data to identify the root cause of the problem. For example, user behavior data is combined with network performance data to analyze changes in user browsing time and purchase conversion rates during periods of increased network latency to determine whether network performance issues are directly affecting user experience and marketing conversion effectiveness. At the same time, combined with marketing campaign data, it is examined whether a surge in traffic to relevant pages during specific promotional events has led to problems such as excessive server load.

[0049] Root cause tracing and location: Based on preset network quality assessment rules and business logic, conduct in-depth layer-by-layer analysis of abnormal data, trace the root cause of the problem, and accurately locate it to the specific link or factor.

[0050] For example, through analysis, it was found that the loading speed of a certain page was slow. Further tracing back, it was found that this was because the page referenced a large number of external image resources, and the server where these image resources were located had a high response delay, thus determining that the root cause of the problem was the loading problem of external image resources.

[0051] Based on the results of problem diagnosis and location, combined with the preset optimization strategy library, optimization suggestions for current network quality issues are generated; Based on the root cause of the problem identified by the Problem Diagnosis and Location Module, a corresponding optimization strategy is matched from a pre-set optimization strategy library. This optimization strategy library contains a variety of optimization solutions and measures for different types of problems (such as network performance issues, user behavior guidance issues, marketing campaign design issues, etc.), as well as their applicable scenarios and expected results. For example, for the aforementioned external image resource loading issue, matching optimization strategies may include optimizing the image resource storage location, using CDN acceleration technology, and compressing the image.

[0052] When multiple optimization solutions exist for a given problem, a comprehensive evaluation and comparative analysis of the matching optimization strategies is conducted. Factors considered include implementation difficulty, cost, expected results, and impact on existing services. Ultimately, a prioritized set of optimization recommendations is generated, providing network operators with comprehensive and feasible options. For example, when considering server upgrades, the advantages and disadvantages of both directly purchasing hardware upgrades and renting cloud servers are evaluated. The most appropriate option is then recommended based on the company's actual situation and budget.

[0053] Generated optimization suggestions are further refined and specified, providing detailed implementation steps, technical parameter settings, and precautions, so that network operators can directly follow and implement the suggestions. For example, suggestions for optimizing image resource loading include detailed instructions on how to select appropriate image compression tools and formats, how to configure CDN acceleration service parameters, and how to optimize page code to improve image loading efficiency, making the optimization suggestions highly operational and practical.

[0054] Based on the results of the problem diagnosis and location module and combined with the preset optimization strategy library, targeted optimization suggestions are generated. In response to the above-mentioned problems of insufficient server storage space and network bandwidth congestion, the optimization suggestion generation module makes the following suggestions: Server upgrade: Increase server storage space, optimize server hardware configuration, and improve server processing capabilities.

[0055] Network bandwidth optimization: Negotiate with network service providers to temporarily increase bandwidth resources during peak hours, or adjust the existing network architecture to achieve reasonable traffic distribution.

[0056] In this embodiment, by adopting the patented multi-module self-inspection network quality method and system based on the perceived marketing conversion index, the e-commerce platform can timely and accurately understand the network quality status, accurately locate problems, and take effective optimization measures. After a period of use, the platform's network quality has been significantly improved, user complaints have decreased, and marketing conversion effects have been significantly improved. For example, the number of participants in marketing activities has increased compared to before, and the advertising conversion rate has increased, effectively improving the platform's business efficiency and market competitiveness.

[0057] See also Figure 2 As shown, in a second aspect of the present invention, a multi-module self-checking network quality system based on the perceived marketing conversion index is provided, characterized in that it includes: The data collection module is used to collect basic data during network operation, including the network perception degradation index, terminal perception degradation index, and perception marketing portrait index; The perceptual marketing conversion index model construction module is used to analyze and process the collected basic data based on the construction of the perceptual marketing conversion index model to calculate the perceptual marketing conversion index; A network quality evaluation module is used to evaluate the network quality based on the calculated perceived marketing conversion index; The problem diagnosis and location module is used to analyze the problem through problem diagnosis methods and locate the fault location through fault location functions when the network quality assessment result is lower than the preset threshold; The optimization suggestion generation module is used to generate optimization suggestions for current network quality issues based on the results of problem diagnosis and location, combined with the preset optimization strategy library. The above embodiments are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary engineering technicians in this field should fall within the scope of protection determined by the claims of the present invention.

Claims

1. A multi-module self-checking network quality method based on the perceived marketing conversion index, characterized by: The following steps are involved: Collect basic data during network operations, including the network perception degradation index, terminal perception degradation index, and perception marketing portrait index; Based on the preset perception marketing conversion index model, the collected basic data is analyzed and processed to calculate the perception marketing conversion index; Evaluate network quality based on the calculated perceived marketing conversion index; When the network quality assessment result is lower than the preset threshold, the problem is analyzed through problem diagnosis methods and the fault location function is used to locate the fault; Based on the results of problem diagnosis and location, combined with the preset optimization strategy library, optimization suggestions for current network quality issues are generated.

2. The multi-module self-checking network quality method based on the perceived marketing conversion index according to claim 1 is characterized in that: The network perception degradation index is obtained through comprehensive judgment of QoE perception score, number of optical modem restarts, Lan port negotiation rate, intranet device negotiation rate, number of PON and WAN interruptions and received optical power.

3. The multi-module self-checking network quality method based on the perceived marketing conversion index according to claim 1 is characterized in that: The terminal perception degradation index is obtained by comprehensively judging the terminal CPU usage, terminal memory resource occupancy, terminal restart interval and terminal temperature.

4. The multi-module self-checking network quality method based on the perceived marketing conversion index according to claim 1 is characterized in that: The perceptual marketing portrait index is obtained through comprehensive judgment of the number of devices connected to the intranet, the number of smart home devices in the intranet, the number of high-value mobile devices, the number of self-purchased 100M old routers, the number of home routers, the number of home Wi-Fi amplifiers and the number of home Linux servers.

5. The multi-module self-checking network quality method based on the perceived marketing conversion index according to claim 1 is characterized in that: The perception marketing conversion index is obtained by weighted combination of the network perception degradation index, the terminal perception degradation index and the perception marketing portrait index, as follows: in, represents the marketing conversion index, represents the network perception degradation index, represents the terminal perception degradation index, represents the perceived marketing portrait index, 、 and represents the weight, where and = =0.2, .

6. The multi-module self-checking network quality method based on the perceived marketing conversion index according to claim 1 is characterized in that: The perception marketing conversion index model is trained using a machine learning algorithm. The training data includes historical network operation data and corresponding marketing conversion result data, and the model parameters are optimized through continuous iteration.

7. The multi-module self-checking network quality method based on the perceived marketing conversion index according to claim 1 is characterized in that: The network quality is assessed in the following ways: in, Indicates the current network quality score. Indicates the perceived marketing portrait index value calculated in the current cycle. Represents a user Current network transmission rate, Indicates the mapping quality score between the perceived marketing profile index value and the network transmission rate, represents the influence coefficient, represents the error delay function.

Images