New product planning system based on AI intelligent analysis

Abstract

The invention relates to the technical field of product planning, and discloses a new product planning system based on AI intelligent analysis, which comprises a data collection unit, an AI analysis unit, a comprehensive analysis unit and a real-time feedback unit, and is characterized in that the data collection unit integrates multi-source data of industries, e-commerce, consumers and the like to guarantee information comprehensiveness; the AI analysis unit performs standardized processing on data, accurately clusters consumer groups and mines demand characteristics of each group, the comprehensive analysis unit fuses market, consumer groups and frontier technical data, associates industry and demand trends, compares product function advantages and disadvantages, and screens adaptive technologies to improve product competitiveness, and the real-time feedback unit performs real-time feedback when a scheme is adjusted. According to the method, data change and analysis results are presented in real time, rapid decision making is assisted, all units supplement each other, a scientific and efficient new product planning system is constructed, enterprises can accurately grasp market requirements, and scientificity, accuracy and timeliness of new product planning are greatly improved.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of product planning, and particularly relates to a new product planning system based on AI intelligent analysis. BACKGROUND

[0002] In today's competitive and rapidly changing business environment, new product planning has become a key link for enterprises to maintain market competitiveness and achieve sustainable development. Traditional new product planning methods often rely on the experience and subjective judgment of planners, and have many limitations in market research, consumer group analysis, and competition situation judgment.

[0003] On the one hand, traditional market research methods are difficult to obtain massive and scattered market data comprehensively and timely. Only relying on questionnaire surveys, interviews and other methods, it is difficult to effectively cover multi-dimensional data such as industry dynamics, competitor information, and consumer multi-channel feedback, resulting in inaccurate grasp of market trends. For example, it is difficult to quickly capture real-time consumer evaluations and potential needs of products on social media, as well as market change signals brought by emerging technologies in the industry.

[0004] On the other hand, in terms of consumer group analysis, traditional methods can only make simple demographic divisions and cannot deeply mine the underlying needs and preference characteristics of consumer behavior. There is a lack of systematic analysis of consumer browsing, searching, and purchasing behavior data on e-commerce platforms, making it difficult to accurately divide consumer groups and thus unable to achieve precise positioning and function optimization of products.

[0005] In addition, in terms of competition analysis, the traditional mode has low efficiency in collecting and analyzing information such as the functions and prices of competitor products, and cannot timely discover the competitive advantages and disadvantages of its own products, making it difficult to quickly respond to market changes and develop effective competition strategies. At the same time, in the process of integrating cutting-edge technologies into products, there is a lack of scientific evaluation methods to judge the matching degree of technology and product demand and the potential of product function improvement, which may cause resource waste or miss technology innovation opportunities.

[0006] How to integrate multi-source data with advanced technology to achieve accurate analysis of consumer groups, intelligent judgment of market competition situation, and efficient integration of cutting-edge technology and product demand has become a problem to be solved in the field of new product planning. SUMMARY

[0007] The purpose of the present application is to provide a new product planning system based on AI intelligent analysis, which solves the technical problems raised in the background art.

[0008] The purpose of the present application can be achieved by the following technical solutions:

[0009] A new product planning system based on AI intelligent analysis, comprising:

[0010] Data collection unit: for collecting target data corresponding to the target product;

[0011] AI analysis unit, first standardizing the target data, and then using clustering analysis to divide the consumer groups, which is based on the consumer purchase data to construct a comprehensive data vector, measuring the similarity by Euclidean distance, calculating the clustering center by iteration, obtaining multiple consumer groups, and then analyzing the consumption habits and demand characteristics of each group;

[0012] Comprehensive analysis unit, first, fuse market and consumer group data, correlate industry growth trend and group demand growth trend, calculate correlation; compare the group preference difference of the product and the competitor in each function, judge the competitive advantages and disadvantages, then fuse the front-end technology and the consumer group data, calculate the matching degree of the front-end technology application scene and the group demand scene, filter the high matching degree technology into the product, and analyze its potential to improve the product function;

[0013] Real-time feedback unit, when the product scheme is adjusted by the planner, real-time call the data and results of the comprehensive analysis unit feedback.

[0014] As a further scheme of the present application: the target data includes:

[0015] Data I, industry report and competitor product information:

[0016] Determine the industry range corresponding to the collected target product, and set the data collection period;

[0017] For industry reports, data interfaces or subscription services are established with professional industry consulting agencies to obtain;

[0018] In the industry report, the market size data is marked as M, and the market size growth trend data is G m ;

[0019] The market size growth trend data represents the change rate of the market size in the specified time period;

[0020] For competitor product information, a competitor product information database is established in advance;

[0021] At the same time, in the competitor product information database, the function parameters of the corresponding products of the competitors are marked as F i,j , wherein i represents different competitors, i=1, 2, ……n, n represents the number of competitors, j represents different function dimensions, j=1, 2, ……m. m represents the number of different function dimensions; and the price information of the corresponding products of the competitors is marked as P i , which represents the price of the i-th competitor product;

[0022] Data II, e-commerce platform and social media data:

[0023] For e-commerce platform data, for the target e-commerce platform, use network data collection tools to collect consumer evaluation data on the same type of target product;

[0024] Among them, the network data collection tool follows the legal data collection protocol when collecting data;

[0025] Mark the number of evaluations of the same type of target product on the e-commerce platform as N e , the number of positive comments as N e-good , and the number of negative comments as N e-bad ;

[0026] At the same time, through:

[0027] Determine the positive comment rate R e-good and the negative comment rate R e -bad of the same type of target product on the e-commerce platform;

[0028] At the same time, collect consumer demand keywords and count the frequency of occurrence of each type of demand keyword FD k2 , where k2 represents different demand keywords;

[0029] For social media data, monitor the social media platform and mark the number of hot topics monitored as H, and the number of hot topics related to the target product field as H r ;

[0030] At the same time, collect consumer feedback information, including complaints and expectations about existing target products, and count the number of occurrences of each type of feedback CF l , where l represents different feedback types;

[0031] Data III, consumer purchase data:

[0032] Cooperate with e-commerce platforms or sales systems to obtain consumer browsing, search, and purchase record data;

[0033] Mark the number of times a consumer browses the same type of target product in the browsing record as B c , the number of times a consumer searches for the same type of target product-related features in the search record as S c , and the number of times a consumer purchases the same type of target product in the purchase record as P c , where c is an index used to identify "consumer" and refers to the cth consumer;

[0034] For the c-th consumer's preference feature for the j-th function of similar target products, extract the number of times the c-th consumer browses similar target products that include the j-th function. c，j The number of searches S in the search history for the same type of target product that includes the j-th function. c，j The number of times P in the purchase history of the same type of target product that includes the j-th function. c，j Simultaneously, extract the total number of views (B) of the c-th consumer for the same type of target product. c，tot The total number of searches for similar target products in the search history, S c，tot The total number of times P, representing the purchase of the same type of target product, in the purchase records. c，tot ;

[0035] pass

[0036] Determine the degree of consumer preference (PR) for the j-th function of similar target products. c,j ;

[0037] In the formula, α1, α2, and α3 are pre-defined weight coefficients, and satisfy α1 + α2 + α3 = 1;

[0038] Data 4: Patent Information and Technical Papers

[0039] For patent information, access the patent database;

[0040] The number of patents collected that are related to the target product is labeled as N. p The patent's technology classification is T p,mc ;

[0041] Where p is an index used to identify "patents", referring to the p-th patent, and mc represents different technology categories;

[0042] Simultaneously, extract the technical features from the patent, and mark the technical feature of the p-th patent as TE. p,k1 SC and its application scenarios p,k1 The frequency of occurrence F(SC) in different application scenarios k1 ;

[0043] Where k1 is the technical feature number;

[0044] For technical papers, collect technical papers in relevant fields from academic databases;

[0045] The number of technical papers collected that are related to the target product is denoted as N. t At the same time, the descriptions of technological development trends involved in technical papers are marked as TR. t,s And extracting application scenarios of new technologies in papers to predict PRR.t,s ;

[0046] Where t is the index used to identify the "technical paper", referring to the t-th technical paper, and s is the trend description number.

[0047] As a further aspect of the present invention, the standardization process is as follows:

[0048] Taking any dimension of target data as an example, and labeling the target data as XM, the standardized value is XM` = XMB = XM-XMP;

[0049] Where XMP is the mean of the target data in this dimension, and XMB is the standard deviation of the target data in this dimension.

[0050] As a further aspect of the present invention, the consumer group is segmented as follows:

[0051] Extract the pre-defined number of cluster categories, Ka;

[0052] The collected consumer purchase data is organized to obtain a consumer group dataset XD = {XD1, XD2, ..., XD}. m0}, j0 = 1, 2, ..., m0, where m0 represents the data dimension, which includes data corresponding to the dimensions of browsing, searching, purchasing, and preferences;

[0053] The similarity between consumer data is measured using the Euclidean distance formula;

[0054] Its formula is

[0055] In the formula, x = (x j0 ) = (x1, x2, ... x m0 ) and y = (y j0 ) = (y1, y2, ... y m0 ) refers to two data vectors consumed;

[0056] Clustering process:

[0057] Randomly select Ka initial cluster centers C s0 ={C s0，j0}={C s0，1 C s0，2 ...C s0，m0}, s0 = 1, 2, ..., Ka, C s0 The data vector representing the s0th cluster center;

[0058] For each consumer data XD j0 Calculate its distance to each cluster center C s0 Euclidean distance D(XD) j0 Cs0 ), and XD j0 is allocated to the population corresponding to the nearest cluster center;

[0059] Then, the new center of each cluster population is recalculated, and each dimension value of the new center is the average of the corresponding dimension values of all consumer data in the population;

[0060] That is, for the s0th population, the data set of the consumer population contained therein is marked as XD ie,j0 , ie=1, 2, ……n0 s0 , n0 s0 is the number of consumers in the s0th population, ie is the index variable of the consumer, ie∈s0, and XD ie,j0 represents the j0th dimension value of the ie consumer data;

[0061] By: determining the j0th dimension value C s0 ' of the new center C s0,j0 ';

[0062] The above allocation and center recalculation steps are repeated until the cluster center no longer changes significantly;

[0063] Wherein, the cluster center no longer changes significantly refers to the change amount of all cluster centers in two iterations is less than a pre-set threshold Y b ;

[0064] The change amount is measured by calculating the square root of the sum of squares of the dimension differences between the new and old centers, that is, When all CC s0 < Y b , stop;

[0065] After the division is completed, s0 consumer populations G1, G2, ……G s0 are obtained;

[0066] Then, the consumption habits and demands of each population are analyzed;

[0067] The consumption habit features and demand features of the s0th population are extracted, and are marked as CH s0，u0 and DE s0，u0 respectively, wherein u0 is the habit feature number.

[0068] As a further scheme of the application, the market and consumer population fusion analysis method is as follows:

[0069] Wherein, the market data refers to industry reports, corresponding data of competitors;

[0070] The industry market size growth trend G mWith consumer group G s0 Demand growth trend GD s0 Perform association analysis:

[0071] in,

[0072] DE s0,t and DE s0,t+1 These represent the demand quantities of the S0th group for the relevant demand characteristics during time period t and time period t+1, respectively.

[0073] The relationship between the two is then analyzed to determine whether the growth trend of consumer demand matches the growth trend of market size. The formula is as follows:

[0074]

[0075] Where T is the number of time periods, GP m It is all G within T time periods m The mean of GDP s0 It is all GD within T time periods s0 The mean of the two, Corr represents the correlation between the two, and G m (t) and GD s0 (t) represents the industry market size growth trend and demand growth trend during the t-th time period, respectively.

[0076] As a further aspect of the present invention, the method for determining competitive advantages and disadvantages is as follows:

[0077] Extract the functional parameters of the target product in this plan as Four j And the competitor's product functional parameters are F i,j ;

[0078] By: PRC s0，i，j =PR our，s0，j -PR i，s0，j

[0079] Identify consumer group G s0 The difference in preference between this product and competitors' products is PRC. s0，i,j ;

[0080] Among them, PR our,s0,j It is consumer group G s0 The degree of preference for the j-th function corresponding to the target product in this plan, PR i,s0,j It represents the degree of preference for the j-th function of a competitor's target product;

[0081] Subsequently, by comparing PRC s0，i,j The sign and magnitude of the value are used to determine the competitive advantages or disadvantages of this product in each functional dimension.

[0082] When PRC s0，i,j is positive, it means that the target product of the present planning is in a competitive advantage on the jth function, that is, more popular with G s0 group preference;

[0083] When PRC s0，i,j is negative, it means that the target product of the competitor is in a competitive advantage on the jth function.

[0084] As a further scheme of the present application: PRC our,s0,j By analyzing the browsing, searching and purchasing behavior data of the consumer group G s0 on the jth function of the product;

[0085] The calculation formula is as follows:

[0086]

[0087] Wherein, β1, β2, β3 are the preset weight coefficients corresponding to the browsing behavior, searching behavior and purchasing behavior respectively, Tbr s0,j is the total browsing time of the G s0 group on the jth function related content, Tbr s0,tot is the total browsing time of the G s0 group on all function related content of the product; Nsea s0,j is the search times of the G s0 group on the jth function, Nsea s0,tot is the total search times of the G s0 group on all functions of the product; Nbuy s0,j is the purchase times of the G s0 group due to the jth function, Nbuy s0,tot is the total purchase times of the G s0 group on the product;

[0088] And the calculation method of PRC i,s0,j is consistent with that of PRC our,s0,j .

[0089] As a further scheme of the present application: the fusion analysis method of the frontier data and the consumer group data is as follows:

[0090] Wherein, the frontier data refers to the data corresponding to the patent information and technical papers;

[0091] Analyze the matching degree of the application scene of the frontier technology and the demand of the consumer group;

[0092] Select a frontier technology, mark its application scene as SC, and obtain the consumer group G s0The demand scenario is marked as DE (SC) s0 ;

[0093] By:

[0094] The matching degree M1 of the application scenario of the frontier technology and the demand of the consumer group is determined s0，SC ;

[0095] The number of overlapping features refers to the number of same or related features in the technology application scenario SC and the consumer group demand scenario DE (SC) s0 , which is obtained by manual marking and statistics.

[0096] As a further scheme of the application: the product function improvement potential analysis is: the matching degree higher than the preset matching degree threshold is integrated into the target product of the planning, and then the matching degree higher than the matching degree threshold is analyzed for the potential of the product function and performance improvement;

[0097] The formula is:

[0098] Wherein, LM j，tech The target product is the jth function improvement degree of the corresponding frontier technology, the new function index is the index that the product function j can reach after applying the technology, and the original function index is the index without applying the technology.

[0099] As a further scheme of the application: the feedback mode of the real-time feedback unit is as follows:

[0100] If the planning personnel adjusts the related function parameters of the target product, the related analysis indexes are immediately recalculated, that is, the difference between the preference degree of the consumer group for the jth function of the same type target product before and after the adjustment, and the difference between the correlation of the growth trend of the consumer group demand and the market size trend before and after the adjustment; Then the recalculated analysis results and suggestions are displayed to the planning personnel in an intuitive way.

[0101] The change of the matching degree of the change of the consumer group preference and the market data;

[0102] It is obtained by comparing the adjusted function parameters with the original parameters, that is, the difference between the preference degree of the consumer group for the jth function of the same type target product before and after the adjustment, and the difference between the correlation of the growth trend of the consumer group demand and the market size trend before and after the adjustment; Then the recalculated analysis results and suggestions are displayed to the planning personnel in an intuitive way.

[0103] The beneficial effects of the application are:

[0104] The application, the system integrates industry reports, competitor information, e-commerce and social media data, consumer purchase data, patents and technical papers, etc. Multidimensional target data is collected by the data collection unit, breaking the data island, realizing the centralized and systematic management of data.

[0105] The application, by means of the AI analysis unit, standardizes the target data, eliminates the format difference and dimension influence of data from different sources, and provides a unified and reliable data basis for subsequent analysis. At the same time, the clustering analysis is used to divide the consumer groups, the comprehensive data vector is constructed based on the consumer purchase data, the similarity is accurately measured by the Euclidean distance, and multiple consumer groups are obtained through iterative calculation, which can deeply mine the consumption habits and demand characteristics of each group, and greatly improve the accuracy and depth of analysis compared with traditional manual analysis.

[0106] The application, the comprehensive analysis unit fuses market and consumer group data, and by correlating the industry growth trend with the group demand growth trend and calculating the correlation, the matching degree of the consumer group demand growth and the market size growth can be clearly judged, helping the planners to grasp the market dynamics and find the market basis of product positioning.

[0107] The application, by comparing the group preference difference of the product and the competitor in each function, can objectively and quantitatively judge the competitive advantages and disadvantages of the product in each function dimension, avoid the deviation of subjective judgment in traditional competitive analysis, provide a clear direction for product optimization, and help enterprises to develop targeted competitive strategies.

[0108] The application, the comprehensive analysis unit fuses the leading technology and the consumer group data, calculates the matching degree of the application scene of the leading technology and the group demand scene, and can filter out the technology with high matching degree. This data-based technology screening method ensures that the introduced leading technology really meets the needs of consumers and avoids the waste of resources caused by blind application of technology.

[0109] The application, by simultaneously analyzing the improvement potential of the leading technology on the product function, can predict the enhancement effect of the technology application on the product competitiveness in advance, providing a scientific basis for product innovation and helping enterprises to create new products with technical advantages and market competitiveness.

[0110] The application, the real-time feedback unit can call the data and results of the comprehensive analysis unit in real time when the planners adjust the product scheme, so that the planners can timely understand the influence of the adjustment on the consumer group preference, market matching degree, competitive advantages and disadvantages, etc.

[0111] The application, the feedback results are presented in an intuitive way, which is convenient for the planners to make decisions quickly, shortens the optimization cycle of the product scheme, improves the planning efficiency, and makes the product scheme dynamically adjusted according to the actual situation, which is more suitable for market changes and consumer needs.

[0112] In the data collection process, the application clearly stipulates that for the e-commerce platform data, the network data collection tool is used to collect data in accordance with the legal data collection protocol, ensuring the legality of the data source, reducing the legal risks of enterprises caused by improper data collection, and protecting the compliance of system operation.

[0113] In summary, the present application realizes the intelligent and accurate whole process of data processing, group analysis, market competition research and judgment, technology fusion and scheme optimization in the new product planning process through AI intelligent analysis technology, effectively improves the scientificity, efficiency and success rate of new product planning, and provides strong support for enterprises to occupy an advantageous position in market competition. BRIEF DESCRIPTION OF DRAWINGS

[0114] The present application will be further described below in conjunction with the accompanying drawings.

[0115] Figure 1 is a system block diagram of a new product planning system based on AI intelligent analysis. DETAILED DESCRIPTION

[0116] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0117] As an embodiment of the present application:

[0118] Please refer to Figure 1 The present application is a new product planning system based on AI intelligent analysis, which comprises:

[0119] Data collection unit: for collecting target data corresponding to the target product;

[0120] The target data comprises:

[0121] Data I, industry report and competitor product information:

[0122] Determine the industry range corresponding to the target product collected, and set the data collection period;

[0123] In this embodiment, for example, the smart watch is specified as a consumer electronics product field, and the update is performed monthly;

[0124] For industry reports, data interfaces or subscription services are established with professional industry consulting agencies to obtain;

[0125] In the industry report, the market size data is marked as M, and the market size growth trend data is G m ;

[0126] Wherein, the market size growth trend data represents the change rate of market size in a specified time period, and the statistical dimension of market size data is divided according to any one of region, user group, and sales channel;

[0127] For example: the monthly growth trend is represented as Wherein, M t is the market size of the t-th month, M t+1 is the market size of the (t+1)-th month, and t is a parameter used to identify the month, representing the specific month number; t+1

[0128] For competitor product information, a competitor product information database is established in advance;

[0129] Meanwhile, in the competitor product information database, the function parameters of the corresponding products of the competitors are marked as F i,j , wherein i represents different competitors, i = 1, 2, … n, n represents the number of competitors, j represents different function dimensions, j = 1, 2, … m, and m represents the number of different function dimensions; and the price information of the corresponding products of the competitors is marked as P i , representing the price of the i-th competitor product;

[0130] In this embodiment, the function dimensions include but are not limited to endurance, screen resolution, etc.;

[0131] Data II: E-commerce platform and social media data:

[0132] For e-commerce platform data, for the target e-commerce platform, the evaluation data of consumers on the same type of target product is collected by using a network data collection tool;

[0133] Wherein, the network data collection tool is implemented based on the Scrapy framework of Python, and the data collection complies with a legal data collection protocol;

[0134] The number of evaluations of the same type of target product on the e-commerce platform is marked as N e , the number of favorable evaluations is N e-good , and the number of unfavorable evaluations is N e-bad ;

[0135] Meanwhile, by:

[0136] The favorable evaluation rate R e-good and the unfavorable evaluation rate R e-bad of the same type of target product on the e-commerce platform are determined;

[0137] Meanwhile, the consumer demand keywords are collected, and the frequency of occurrence of various demand keywords FD k2 ​, wherein k2 represents different demand keywords;

[0138] In this embodiment, the demand keywords FD k2 are extracted by any one of the word segmentation, stop word removal, and semantic clustering algorithm in natural language processing;

[0139] In this embodiment, the consumer demand keywords are, for example, “hope that the product has longer battery life” and the like;

[0140] For social media data, the social media platform is monitored, and the number of monitored hot topics is marked as H, and the number of hot topics related to the target product field is H r ;

[0141] In this embodiment, the determination criterion of the hot topic H r is that the topic reading volume and the discussion volume correspond to a preset threshold value, that is, the topic is determined to be a hot topic only when it exceeds the corresponding threshold value;

[0142] Meanwhile, consumer feedback information is collected, which includes complaints and expectations about the existing target product, and the occurrence frequency CF l of each type of feedback is counted, wherein l represents different feedback types;

[0143] In this embodiment, the social media platform is, for example, a social network, a forum, and the like;

[0144] Data three, consumer purchase data:

[0145] Cooperate with e-commerce platforms or sales systems to obtain consumer browsing, searching, and purchase record data;

[0146] The browsing frequency of the same type of target product in the consumer browsing record is marked as B c , the searching frequency of the same type of target product related features in the search record is marked as S c , and the purchase frequency of the same type of target product in the purchase record is marked as P c , and c is an index for identifying “consumer”, which refers to the cth consumer;

[0147] In this embodiment, the searching frequency of the same type of target product related features in the search record is, for example, the searching frequency of “smart watch waterproof”;

[0148] For the preference feature of the jth function of the same type of target product of the cth consumer, the browsing frequency B c，j of the cth consumer on the same type of target product covering the jth function, the searching frequency S c，j of the same type of target product covering the jth function in the search record, and the purchase frequency Pc，j Meanwhile, the total number of times of browsing the same type of target product by the cth consumer B is extracted c，tot The total number of times of searching the same type of target product in the search record S is searched c，tot The total number of times of purchasing the same type of target product in the purchase record P is searched c，tot ;

[0149] By

[0150] The preference degree of the consumer to the jth function of the same type of target product is determined as PR c,j ;

[0151] In the formula, a1, a2, and a3 are preset weight coefficients, and satisfy a1+a2+a3=1;

[0152] In this embodiment, the values of a1, a2, and a3 are determined according to actual data statistics and industry experience, and the initial values can be set as a1=0.3, a2=0.3, and a3=0.4, which can be adjusted according to verification subsequently;

[0153] Data four, patent information and technical papers:

[0154] For patent information, access the patent database;

[0155] In this embodiment, the accessed patent database follows a legal access channel;

[0156] The number of patents collected in the field related to the target product is marked as N p The technical classification of the patent is T p,mc ;

[0157] Wherein, p is an index for identifying “patent”, which refers to the pth patent, and mc represents different technical categories, such as battery technology, sensor technology, etc.

[0158] Meanwhile, the technical features in the patent are extracted, and the technical features of the pth patent are marked as TE p,k1 The description of the application scenario SC p,k1 And the frequency F(SC) of different application scenarios k1 ;

[0159] Wherein, k1 is the technical feature number, and the extraction rule of the technical feature TE p,k1 is determined based on the keyword extraction of the patent claim and the association of the technical classification number, and the extraction rule of the description of the application scenario SC p,k1 is based on the keyword extraction of the specific implementation manner of the patent specification and the association of the technical classification number;

[0160] For technical papers, collect relevant technical papers in the field from academic databases;

[0161] In this embodiment, the use of academic databases complies with the legal use agreement;

[0162] Mark the number of technical papers collected as N t , and mark the technical development trend described in the technical paper as TR t,s , and extract the application scenario prediction PRR of new technology in the paper t,s ;

[0163] Where t is the index for identifying "technical papers", which refers to the tth technical paper, and s is the trend description number;

[0164] The AI analysis unit first standardizes the target data, and then uses clustering analysis to divide the consumer groups. The method is to construct a comprehensive data vector based on consumer purchase data, measure similarity by Euclidean distance, calculate the clustering center by iteration, obtain multiple consumer groups, and then analyze the consumption habits and demand characteristics of each group.

[0165] The specific method is as follows:

[0166] First, standardize the target data;

[0167] Select any dimension of target data as an example, and mark its target data as XM, and the standardized value as XM`=XMB×XM-XMP

[0168] Where XMP is the mean of the target data in this dimension, and XMB is the standard deviation of the target data in this dimension.

[0169] In this embodiment, standardization processing makes different dimension data have the same order of magnitude in subsequent analysis;

[0170] Use clustering analysis method to construct and divide consumer groups:

[0171] Extract the pre-set number of clustering categories Ka;

[0172] Where the value of Ka can be initially set according to actual data and industry experience, and then verified and adjusted;

[0173] Organize the collected consumer purchase data, and get the data set of consumer groups XD={XD1、XD2、……XD m0}, Where XD i0 represents the comprehensive data vector of the ith consumer, which includes the data of the corresponding dimensions of browsing, searching, purchasing and preference;

[0174] The similarity between the consumer data is measured by the Euclidean distance formula;

[0175] The formula is

[0176] In the formula, x=(x1, x2, … x m0 ) and y=(y1, y2, … y m0 ) refer to two consumer data vectors, j0=1, 2, … m0, and m0 refers to the data dimension;

[0177] Clustering process:

[0178] Randomly select Ka initial cluster centers C s0 ={C s0，j0}={C s0，1 , C s0，2 , … C s0，m0}, s0=1, 2, … Ka, C s0 represents the data vector of the s0th cluster center;

[0179] For each consumer data XD j0 , calculate its Euclidean distance D(XD s0 , C j0 ) to each cluster center C s0 , and assign XD j0 to the group corresponding to the nearest cluster center;

[0180] Then recalculate the new center of each cluster group, and the value of each dimension of the new center is the average value of the corresponding dimension values of all consumer data in the group;

[0181] That is, for the s0th group, mark the data set of the consumer group contained therein as XD ie,j0 , ie=1, 2, … n0 s0 , n0 s0 is the number of consumers in the s0th group, ie is the index variable of the consumer, ie∈s0, and XD ie,j0 represents the j0th dimension value of the ie consumer data in the s0th group;

[0182] By: determine the j0th dimension value C s0,j0 ' of the new center C s0 ';

[0183] Repeat the above assignment and center recalculation steps until the cluster centers no longer change significantly;

[0184] where the cluster centers no longer change significantly means that the change amount of all cluster centers in two iterations is less than a pre-set threshold Yb ;

[0185] In this embodiment, Y b = 0.01;

[0186] The change amount is measured by calculating the square root of the sum of squares of the differences in each dimension of the new and old centers, that is, When all CC s0 < Y b , stop;

[0187] After the division is completed, s0 consumer groups G1, G2, …, G s0 are obtained.

[0188] Then the consumption habits and demands of each group are analyzed;

[0189] The consumption habit features and demand features of the s0th group are extracted and are marked as CH s0，u0 and DE s0，u0 , respectively, where u0 is the habit feature number.

[0190] In this embodiment, the consumption habit features are, for example, purchase frequency, purchase time period, etc. The division criteria of the "purchase time period" can be, for example, division according to weekdays and weekends, daytime and nighttime, etc. The demand features are, for example, demand for product functions, price, etc., which are obtained by statistics of the consumer data in the group.

[0191] Comprehensive analysis unit: first, fuse market and consumer group data, correlate industry growth trend and group demand growth trend, calculate correlation; compare the group preference difference of the product and the competitor in each function, judge the competitive advantages and disadvantages, then fuse the leading technology and the consumer group data, calculate the matching degree of the leading technology application scene and the group demand scene, filter the high matching degree technology into the product, and analyze its potential for improving product functions;

[0192] The specific way of fusing and analyzing market data and consumer group data is as follows:

[0193] The market data refers to the data corresponding to the industry report and the competitor;

[0194] Correlate the industry market size growth trend G m with the demand growth trend GD s0 of the consumer group G s0 :

[0195] Wherein,

[0196] DE s0,t and DE s0,t+1is the number of demand for the relevant demand characteristics in the tth time period and the t+1th time period, respectively;

[0197] Then, the relationship between the two is analyzed to determine whether the demand growth trend of the consumer group matches the market size trend growth, and the formula is:

[0198]

[0199] Where T is the number of time periods, GP m is the average of all G m in the T time periods, GDP s0 is the average of all GD s0 in the T time periods, and Corr represents the correlation between the two; G m (t) and GD s0 (t) are the market size growth trend and the demand growth trend in the tth time period, respectively.

[0200] In this embodiment, the value of Corr ranges from -1 to 1, close to 1 indicating a high positive correlation, close to -1 indicating a high negative correlation, and 0 indicating no correlation.

[0201] At the same time, combined with the product information of the competitors, the preference difference of the consumer group for the target product of the plan and the products of the competitors is analyzed.

[0202] The functional parameters of the target product of the plan are extracted as Four j , and the functional parameters of the competitor products are F i,j .

[0203] Through: PRC s0，i，j = PR our，s0，j - PR i，s0，j

[0204] The preference difference of the consumer group G s0 for the target product and the competitor products is determined as PRC s0，i,j .

[0205] PR our,s0,j is the preference degree of the consumer group G s0 for the jth function of the target product of the plan; and PR i,s0,j is the preference degree for the jth function of the target product of the competitor.

[0206] Then, by comparing the positive and negative and size of PRC s0，i,j , the competitive advantage or disadvantage of the product in each functional dimension is determined.

[0207] When PRC s0，i,jIf the value is positive, it means that the target product of the plan is more attractive than the competitor's target product in the jth function. s0 Group preference;

[0208] When the PRC s0，i,j value is negative, it means that the competitor's target product is more advantageous in the jth function.

[0209] The specific way of fusing and analyzing the frontier data and consumer group data is as follows:

[0210] Frontier data refers to the data corresponding to patent information and technical papers;

[0211] Analyze the matching degree of the application scenario of the frontier technology and the demand of the consumer group;

[0212] Select a frontier technology, mark its application scenario as SC, and at the same time obtain the demand scenario of the consumer group G s0 , and mark it as DE(SC) s0 ;

[0213] By:

[0214] Determine the matching degree M1 s0，SC of the application scenario of the frontier technology and the demand of the consumer group;

[0215] Among them, the number of overlapping features refers to the number of same or related features in the technology application scenario SC and the consumer group demand scenario DE(SC) s0 , which is obtained by manual annotation and statistics;

[0216] According to the matching degree result, filter out the frontier technology suitable for being integrated into the new product plan, that is, integrate the frontier technology corresponding to the matching degree higher than the preset matching degree threshold into the target product of the plan;

[0217] Then, for the frontier technology with a matching degree higher than the matching degree threshold, analyze its potential for improving product functions and performance;

[0218] The analysis method is:

[0219] By:

[0220] Determine the improvement degree LM j，tech of the technology to the jth function of the target product;

[0221] Among them, the new function indicator is the indicator that the product function j can reach after applying the technology, and the original function indicator is the indicator without applying the technology;

[0222] Embodiment one makes the data collection more targeted and operable by specifying the industry scope, cycle and data processing details of data collection for specific products such as smart watches. The e-commerce data is collected by using the Scrapy framework of Python, and the demand keywords are extracted by natural language processing to ensure legal collection and efficient analysis of data. In the consumption group division, the standardized processing unifies the data magnitude, and the clustering analysis accurately divides the groups based on the consumer purchase behavior to determine the consumption habits and demand characteristics of each group, providing solid data support for subsequent product planning, realizing the standardized and scientific operation of the whole process from data collection to group analysis, and significantly improving the quality and efficiency of the early preparation work of new product planning.

[0223] As embodiment two of the application:

[0224] Please refer to Figure 1 , compared with embodiment one, the technical solution of the embodiment is only different from embodiment one in that in the embodiment, PR our,s0,j By analyzing the consumption group G s0 , the browsing, searching and purchasing behavior data of the product function dimension j are comprehensively calculated.

[0225] The calculation formula is as follows:

[0226]

[0227] Among them, β1, β2, β3 are respectively the preset weight coefficients corresponding to the browsing behavior, searching behavior and purchasing behavior, Tbr s0,j is the total browsing time of the group G s0 for the function j related content, Tbr s0,tot is the total browsing time of the group G s0 for all function related content of the product; Nsea s0,j is the search times of the group G s0 for the function j, Nsea s0,tot is the total search times of the group G s0 for all functions of the product; Nbuy s0,j is the purchase times of the group G s0 for the function j, Nbuy s0,tot is the total purchase times of the group G s0 for the product;

[0228] And the calculation method of PR i,s0,j is the same as that of PR our,s0,j ;

[0229] Among them, before the preference degree calculation, the data corresponding to the browsing behavior, searching behavior and purchasing behavior are standardized to make different types of data comparable in the interval [0, 1].

[0230] The second embodiment focuses on optimizing the calculation method of product function preference degree. By comprehensively calculating the browsing, searching and purchasing behavior data of the consumer group, and standardizing the data before calculation, the different types of data can be compared in the [0, 1] interval, effectively eliminating the influence of data dimension difference, and improving the accuracy and rationality of preference calculation. This optimization can more accurately reflect the real preferences of the consumer group for the functions of the product, helping enterprises to deeply understand the market acceptance of their products in the function dimension, so as to better meet the needs of consumers in product function design and optimization, and enhance the market competitiveness of the product.

[0231] As the third embodiment of the present application:

[0232] Please refer to Figure 1 The technical scheme of the present application is to combine the schemes of the first embodiment and the second embodiment. The difference between the technical scheme of the present application and the first embodiment and the second embodiment is that the present application further includes:

[0233] Real-time feedback unit: when the planner performs operation or adjustment scheme, real-time call the data used by the comprehensive analysis unit and the results derived therefrom for feedback;

[0234] In the following way:

[0235] If the planner adjusts the function parameters related to the target product, the relevant analysis indicators are immediately recalculated, that is,

[0236] The change in the degree of fit between the change in consumer group preference and market data;

[0237] It is obtained by comparing the adjusted function parameters with the original parameters, that is, the difference between the preference degree of the consumer group for the same type of target product corresponding to the jth function before and after adjustment, and the difference between the correlation of the growth trend of consumer group demand and the growth trend of market size before and after adjustment;

[0238] Then the recalculated analysis results and suggestions are displayed to the planner in an intuitive way;

[0239] For example, when adjusting the product price, according to the price sensitivity of the consumer group, the change in the predicted purchase quantity after adjustment is calculated, and by combining market size data, suggestions about the impact of the price adjustment on market share are given, such as "the current price adjustment makes the predicted purchase quantity of the product in the consumer group G1 increase X%, combined with the growth trend of market size, it is expected to increase market share Y%, but attention should be paid to the product layout of competitors in this price interval, and the measures they may take to affect the market share of the product".

[0240] Wherein, the price sensitivity is calculated by the change of the purchase amount in different price intervals in the historical purchase data, and the formula is:

[0241] Wherein, PSM refers to the price sensitivity, CP buy is the change rate of the purchase amount, and BP is the change rate of the price.

[0242] Embodiment three is based on the integration of the advantages of the first two embodiments, and a real-time feedback unit is added. When the product planning personnel adjusts the product function parameters, the system can real-time recalculate the change of the consumer group preference and the change of the market data, and intuitively display the analysis results and suggestions, such as providing comprehensive suggestions for price adjustment in combination with the price sensitivity and market size data. This real-time feedback mechanism enables the planning personnel to quickly grasp the impact of the scheme adjustment, timely adjust the product planning strategy, realize the dynamic optimization of the product scheme, greatly shorten the iteration cycle of product planning, and improve the response speed of the enterprise to market changes and the flexibility of product planning.

[0243] As embodiment four of the present application:

[0244] Please refer to Figure 1 Compared with embodiments one, two and three, the technical scheme of the present embodiment is to combine the above-mentioned embodiments one, two and three.

[0245] Embodiment four fully integrates the technical schemes of the first three embodiments, integrates data accurate collection and analysis, function preference deep calculation, competitive situation scientific research, front technology efficient fusion and real-time dynamic feedback, and builds a complete and efficient new product planning system. From the data source, the information is ensured to be comprehensive and accurate, the market demand and competitive opportunities are mined through deep analysis, the product scheme is quickly optimized with the help of real-time feedback, the scientificity, accuracy and timeliness of new product planning are comprehensively ensured, the new product that meets market demand and has competitive advantage is provided with strong support, and the core competitiveness of the enterprise in market competition is significantly improved.

[0246] It should be pointed out that: all the data collected in the present application are collected with the consent and authorization of the user, and the use of the data is legal and compliant, and the use and processing of the data comply with the relevant laws, regulations and standards of the relevant region.

[0247] The above formulas are all dimensionless values, and the formulas are obtained by collecting a large amount of data to simulate the most recent real situation, and the preset parameters and threshold values in the formula are set by the person skilled in the art according to the actual situation.

[0248] The above is only a specific embodiment of the present application, but the protection scope of the present application

[0249] Without intending to limit the scope of the application, any changes or modifications

[0250] within the technical scope of the application should be covered by the scope of the

[0251] application. Therefore, the scope of the application should be defined by the scope of

[0252] the claims.

Claims

1. A new product planning system based on AI intelligent analysis, characterized in that, include: The data collection unit is used to collect target data for the target product, including industry reports and competitor product information, such as market size and growth trends. It obtains industry reports through professional agency interfaces or subscriptions and establishes a database to collect competitor product functional parameters and prices. E-commerce platform and social media data includes positive review rates, negative review rates, and frequency of demand keywords for similar target products. It uses the Scrapy framework to collect e-commerce reviews and monitors trending topics and consumer feedback on social media. Consumer purchase data includes consumer browsing, search, purchase records, and functional preference data. Patent and technical paper information includes patent technology features, application scenarios, and technical trends and application predictions related to the target product. The AI ​​analysis unit first standardizes the target data, then uses cluster analysis to divide consumer groups. The method is to construct a comprehensive data vector based on consumer purchase data, measure similarity through Euclidean distance, and iteratively calculate cluster centers to obtain multiple consumer groups, and then analyze the consumption habits and demand characteristics of each group. The comprehensive analysis unit first integrates market and consumer group data, correlates industry growth trends with group demand growth trends, and calculates the correlation. By comparing the differences in group preferences for various functions between this product and its competitors, we can determine the competitive advantages and disadvantages. Then, by integrating cutting-edge technology and consumer group data, we can calculate the matching degree between the application scenarios of cutting-edge technologies and the needs of the group, select technologies with high matching degree to integrate into the product, and analyze their potential to improve product functions. The real-time feedback unit allows planners to access data and results from the comprehensive analysis unit in real time when adjusting product plans.

2. The new product planning system based on AI intelligent analysis according to claim 1, characterized in that, The target data includes: In industry reports and competitor product information, market size data is labeled as M, while market size growth trend data is labeled as G. m Market size growth trend data represents the rate of change in market size over a specified time period. In the competitor's product information database, mark the functional parameters of the corresponding competitor's product as F. i,j Where i represents different competitors, i = 1, 2, ..., n, n represents the number of competitors, j represents different functional dimensions, j = 1, 2, ..., m, m represents the number of different functional dimensions; and the price information of the corresponding products of competitors is marked as P. i , where represents the price of the i-th competitor's product; In e-commerce platform and social media data, the number of reviews for similar target products on e-commerce platforms is labeled as N. e The number of positive reviews is N e-good The number of negative reviews is N e-bad ; Positive review rate (R) of similar target products on e-commerce platforms e-good And negative review rate R e-bad pass It can be concluded that; FD of the frequency of various demand keywords k2 Where k2 represents keywords for different needs; The number of trending topics detected is labeled H, and the number of trending topics related to the target product area is labeled H. r ; The frequency of occurrence of each type of feedback in consumer feedback information is marked as CF. l , where l represents different feedback types, including complaints and expectations about existing target products; In consumer purchase data, the number of times a consumer browses similar target products in their browsing history is marked as B. c The number of searches for features related to the same type of target product in the search history is marked as S. c The number of times the same target product was purchased in the purchase record is marked as P. c c is an index used to identify the "consumer", referring to the c-th consumer; In the patent and technology paper information, the number of patents collected in the field related to the target product is marked as N. p The patent's technology classification is T p,mc p refers to the p-th patent, and mc represents different technology categories; the technical features of the p-th patent are marked as TE. p,k1 And the descriptions of application scenarios and the frequency of occurrence of different application scenarios are respectively marked as SC. p,k1 and F(SC) k1 k1 is the technical feature number; The number of technical papers collected that are related to the target product is denoted as N. t At the same time, the descriptions of technological development trends involved in technical papers are marked as TR. t,s And the application scenario prediction of the new technology in the paper is labeled as PRR. t,s t refers to the t-th technical paper, and s is the trend description number.

3. The new product planning system based on AI intelligent analysis according to claim 2, characterized in that, For the c-th consumer's preference feature for the j-th function of similar target products, extract the number of times the c-th consumer browses similar target products that include the j-th function. c，j The number of searches S in the search history for the same type of target product that includes the j-th function. c，j The number of times P in the purchase history of the same type of target product that includes the j-th function. c，j Simultaneously, extract the total number of views (B) of the c-th consumer for the same type of target product. c，tot The total number of searches for similar target products in the search history, S c，tot The total number of times P, representing the purchase of the same type of target product, in the purchase records. c，tot ; pass Determine the consumer's preference level (PR) for the j-th function of a similar target product. c,j ; In the formula, α1, α2, and α3 are pre-set weighting coefficients, and satisfy α1 + α2 + α3 = 1.

4. The new product planning system based on AI intelligent analysis according to claim 3, characterized in that, The standardized processing method is as follows: Select target data in any dimension and label it as XM. The standardized value is XM` = XMB × XM - XMP. Where XMP is the mean of the target data in this dimension, and XMB is the standard deviation of the target data in this dimension.

5. A new product planning system based on AI intelligent analysis according to claim 4, characterized in that, The consumer groups are segmented as follows: Extract the pre-defined number of cluster categories, Ka; The collected consumer purchase data is organized to obtain a consumer group dataset XD = {XD} j0 } = {XD1, XD2, ..., XD} m0 }, j0 = 1, 2, ..., m0, where m0 represents the data dimension, which includes data corresponding to the dimensions of browsing, searching, purchasing, and preferences; The similarity between consumer data is measured using the Euclidean distance formula; Its formula is In the formula, x = (x j0 ) = (x1, x2, ... x m0 ) and y = (y j0 ) = (y1, y2, ... y m0 ) refers to two data vectors consumed; Clustering process: Randomly select Ka initial cluster centers C s0 ={C s0，j0 }={C s0，1 C s0，2 ...C s0，m0 }, s0 = 1, 2, ..., Ka, C s0 The data vector representing the s0th cluster center; For each consumer data XD j0 Calculate its distance to each cluster center C s0 Euclidean distance D(XD) j0 C s0 ), and XD j0 They are assigned to the group corresponding to the nearest cluster center; Next, the new center of each cluster is recalculated, and the value of each dimension of the new center is the mean of the corresponding dimension values ​​of all consumer data in that cluster; That is, for the s0th group, the dataset of consumer groups contained therein is labeled as XD. ie,j0 ie = 1, 2, ..., n0 s0 n0 s0 Let be the number of consumers in the s0th group, and let ie be the index variable of the consumer, ie∈s0, and XD ie,j0 This is represented as the j0th dimension value of the ieth consumer data in the s0th group; pass: New Center C has been determined. s0 The value of the j0th dimension C s0,j0 '; Repeat the above steps of assigning and recalculating the centers until the cluster centers no longer change significantly. Here, "no longer significantly changing cluster centers" means that the change in all cluster centers during the two iterations is less than a pre-set threshold Y. b ; The change is measured by the square root of the sum of the squares of the differences between the old and new centers in each dimension, i.e. When all CC s0 <Y b Stop at this time; After the segmentation is completed, s0 consumer groups are obtained: G1, G2, ..., G s0 ; Subsequently, the consumption habit characteristics and demand characteristics of the s0th group were extracted and labeled as CH respectively. s0，u0 and DE s0，u0 , where u0 is the custom feature number.

6. A new product planning system based on AI intelligent analysis according to claim 5, characterized in that, The analysis method for integrating the market and consumer groups is as follows: Here, market data refers to industry reports and data corresponding to competitors; The industry market size growth trend G m With consumer group G s0 Demand growth trend GD s0 Perform association analysis: in, DE s0,t and DE s0,t+1 These represent the demand quantities of the S0th group for the relevant demand characteristics during time period t and time period t+1, respectively. The relationship between the two is then analyzed to determine whether the growth trend of consumer demand matches the growth trend of market size. The formula is as follows: Where T is the number of time periods, GP m It is all G within T time periods m The mean of GDP s0 It is all GD within T time periods s0 The mean of the two, Corr represents the correlation between them, and G m (t) and GD s0 (t) represents the industry market size growth trend and demand growth trend during the t-th time period, respectively.

7. A new product planning system based on AI intelligent analysis according to claim 6, characterized in that, The methods for judging competitive advantages and disadvantages are as follows: Extract the functional parameters of the target product in this plan as Four j And the competitor's product functional parameters are F i,j ; Among them, Four j Includes consumer group G s0 Browsing, searching, and purchasing behavior data for the j-th functional dimension of this product; The competitor's product features are F i,j Includes consumer group G s0 Browsing, searching, and purchasing behavior data for the j-th functional dimension of competitors' products; Subsequently passed: Consumer Group G s0 PR (Presentation / Preference) for the j-th function corresponding to the target product in this plan our,s0,j ; Where β1, β2, and β3 are preset weighting coefficients corresponding to browsing behavior, search behavior, and purchase behavior, respectively. s0,j It is G s0 Total browsing time of the group for content related to function j, Tbr s0,tot It is G s0 Total browsing time of the group for all content related to the functions of this product; Nsea s0,j It is G s0 The number of searches for function j by the group, Nsea s0,tot It is G s0 Total number of searches for all features of this product by the group; Nbuy s0,j It is G s0 The number of purchases generated by the group due to function j, Nbuy s0,tot It is G s0 Total number of purchases of this product by the group; At the same time, according to PR our,s0,j The calculation logic calculates the preference level (PR) of the competitor's target product for the j-th function. i,s0,j ; Then via: PRC s0,i,j =PR our,s0,j -PR i,s0,j Identify consumer group G s0 The difference in preference between this product and competitors' products is PRC. s0，i,j ; Subsequently, by comparing PRC s0，i,j The sign and magnitude of the value are used to determine the competitive advantages or disadvantages of this product in each functional dimension. When PRC s0，i,j If the value is positive, it indicates that the target product of this plan has a competitive advantage in the j-th function; When PRC s0，i,j A negative value indicates that the competitor's target product has a competitive advantage in the j-th function.

8. A new product planning system based on AI intelligent analysis according to claim 7, characterized in that, The method for integrating cutting-edge data and consumer group data is as follows: Among them, cutting-edge data refers to the data corresponding to patent information and technical papers; Analyze the matching degree between the application scenarios of cutting-edge technologies and the needs of consumer groups; Select a cutting-edge technology, label its application scenario as SC, and simultaneously obtain the consumer group G. s0 The required scenarios are labeled as DE(SC). s0 ; pass: Determine the degree of match between the application scenarios of cutting-edge technologies and the needs of consumer groups (M1). s0，SC ; The number of overlapping features refers to the number of technology application scenarios (SC) and consumer group demand scenarios (DE(SC)). s0 The number of identical or related features is obtained through manual annotation and statistics.

9. A new product planning system based on AI intelligent analysis according to claim 8, characterized in that, The analysis of the product's functional improvement potential is as follows: cutting-edge technologies with matching degrees higher than the preset matching degree threshold are integrated into the target product of this plan. Then, for cutting-edge technologies with matching degrees higher than the matching degree threshold, their potential to improve product functions and performance is analyzed. The formula is: Among them, LM j，tech This refers to the degree to which the corresponding cutting-edge technology enhances the j-th function of the target product. The new function indicator is the indicator that the product function j can achieve after applying the technology, while the original function indicator is the indicator when the technology is not applied.

10. A new product planning system based on AI intelligent analysis according to claim 9, characterized in that, The feedback method of the real-time feedback unit is as follows: If the planners adjust the relevant functional parameters of the target product, the relevant analysis indicators should be recalculated immediately. Changes in the alignment between changes in consumer preferences and market data; It is obtained by comparing the adjusted functional parameters with the original parameters, namely the difference between the degree of preference of the consumer group for the j-th function of the same type of target product before and after the adjustment, and the difference between the correlation between the growth trend of consumer group demand and the growth trend of market size before and after the adjustment. The recalculated analysis results and recommendations will then be presented to the planners in a visually intuitive way.

Images