Research On The Construction And Application Of A Data-driven Product Prediction Neural Network For Public Platforms

In the era of digitalization,intelligence,and networking,facing the market environment of fierce competition and serious product homogeneity,on the one hand,in order to solve the problems of difficult data acquisition,small amount of data,and poor timeliness of data value in the traditional product design process To achieve efficient and accurate conversion of massive network data into valuable information,accurately grasp user needs and user pain points and then optimize the design,on the other hand,to scientifically screen and evaluate design solutions,and establish a market-oriented consumption The mapping relationship between user perception and product expression helps enterprises predict the user satisfaction feedback evaluation of product design schemes,and proposes a product design method based on a public platform datadriven product prediction neural network model to drive product innovation with big data The design is combined with the new generation of neural network artificial intelligence technology,and introduced into the product design methodology,aiming to optimize product design and build a neural network nonlinear mapping model of user satisfaction,improve the success rate of products in the market,and avoid design risks.Taking a hair dryer as a design case,research is conducted from two aspects: big datadriven product design and design scheme neural network prediction model construction.Big data-driven product design,converting a large amount of data into valuable information,extracting product design elements and guiding product design.The case design process takes user needs as the starting point,using data crawler technology to obtain user evaluation data of hair dryers on the JD e-commerce platform,through de-duplication,word separation and other data processing techniques,TF-IDF keyword acquisition technology,to obtain effective product vocabulary and product sentiment analysis table,mining the potential information value of the data,analyzing data to acquire hair dryer design elements with high user attention and user pain points,extracting hair dryer optimization direction,and guiding new hair dryer design.The neural network prediction model establishes the mapping relationship between hair dryer product design elements and user satisfaction through the known hair dryer product data,and performs model training and learning on the sample data on the MATLAB platform to complete the construction of the BP neural network model and further predict the new hair dryer.The user satisfaction score after the development and design scheme is put into the market is used to evaluate new schemes,and it can also help enterprises to choose the optimal design scheme among many schemes.The following conclusions can be drawn during the research:(1)Through the hair dryer case,the feasibility of the research method and research theory is verified;(2)From the two perspectives of users and enterprises,the product design method of "big data drive" + "intelligent neural network prediction model" is proposed,which can guide the design process in the forward direction and make the design plan more scientific.In the reverse direction,it can predict the user Satisfaction evaluation value,which evaluates the design scheme,is more innovative and economically valuable to the enterprise;(3)The research results are an important expansion and effective supplement to the product innovation design methodology,which improves the new product development and design process and proposes a new idea and design method;(4)This model is suitable for the application of hair dryers and small household appliances,and it also serves as a guide and reference for the development and design of other products;(5)The future research prospects are relatively broad,and the breadth and depth of research can be expanded to make more comprehensive research.