Neural Network Purchasing Behavior Analysis with Distributed Representation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for analyzing purchasing behavior on social networking services lack accuracy in determining user interest, desire, and future purchasing likelihood, as they fail to effectively utilize advanced machine learning techniques to process and interpret user-posted data.
Innovation Solution
A purchasing behavior analysis apparatus employing an artificial neural network with three determination layers to convert social networking service posting information into distributed representations, calculating interest, desire, and purchase likelihood probabilities, allowing for precise phase determination of user purchasing behavior.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional methods are used to analyze purchasing behavior on social networking services, then the analysis process is simple, but the accuracy in determining user interest, desire, and future purchasing likelihood is low
Solution Approach 1:
The patent segments the purchasing behavior analysis into three distinct determination layers within the artificial neural network: interest determination layer, desire determination layer, and purchase likelihood determination layer. Each layer processes specific aspects of user behavior independently, allowing for precise measurement of different purchasing phases while maintaining manageable system complexity through modular architecture.
Solution Approach 2:
The patent introduces distributed representation as an intermediary layer between the input posting information and the determination layers. This intermediary converts unstructured social media text into structured numerical representations, enabling the neural network to process and analyze user behavior with high accuracy without requiring direct complex processing of raw text data.
2Measurement precision
If advanced machine learning techniques are employed to process user-posted data, then the accuracy of purchasing behavior determination is improved, but the computational complexity and processing requirements increase
Solution Approach 1:
The computational workload is segmented across three specialized determination layers, each focused on a specific aspect of purchasing behavior (interest, desire, likelihood). This segmentation allows the system to process data efficiently by dedicating specific computational resources to each determination task, reducing overall processing requirements compared to a single monolithic model.
Solution Approach 2:
The patent performs preliminary conversion of posting information into distributed representations before inputting data into the determination layers. This pre-processing step transforms unstructured text into optimized numerical formats, reducing the computational burden on subsequent determination layers and enabling accurate analysis with lower processing power requirements.
Data Source
AI summary
A purchasing behavior analysis apparatus includes an acquiring unit that acquires posting information about a specific product from posting information posted to a social networking service; a conversion unit; and an artificial neural network. The artificial neural network includes a first determination layer which determines whether a user is interested in the product, a second determination layer which determines whether the user wants the product, and a third determination layer which determines whether the user is predicted to purchase the product in the future. The purchasing behavior analysis apparatus further includes an interest presence probability calculating unit, a purchase desire probability calculating unit, and a purchase likelihood probability calculating unit.