File Path Predictive Model for Collaboration Recommendations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing collaboration systems face challenges in presenting relevant content object recommendations to users, often resulting in overwhelming numbers of irrelevant suggestions due to assumptions about shared interests, leading to low user engagement.
Innovation Solution
The development of techniques that utilize file path information to form a predictive model, reducing resource consumption and improving recommendation relevance by determining content objects a user is likely to access, based on historical access activity and file path attributes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If collaboration recommendations are formed based on assumed shared interests between users, then the number of recommendations presented to users increases, but the relevance of recommendations decreases and user engagement drops
Solution Approach 1:
The system performs preliminary analysis of file path information and user access patterns before generating recommendations. By pre-processing historical access data and establishing predictive models in advance, the system can quickly determine relevant recommendations without relying on broad assumptions about shared interests, thus maintaining high relevance while managing the quantity of recommendations effectively
Solution Approach 2:
The system changes the parameters used for recommendation generation from user relationship-based metrics to file path-based metrics. Instead of using user department, user role, or collaboration history as parameters, the system uses file path attributes (such as directory structure, file type, access patterns) as the basis for determining recommendation relevance, which fundamentally improves recommendation quality
2Quantity of substance
If comprehensive user relationship data is used to generate recommendations, then more recommendations can be presented, but system resource consumption increases
Solution Approach 1:
The system extracts and uses only the necessary file path information from the vast amount of available user relationship data. Instead of processing comprehensive user relationship graphs, collaboration histories, and organizational structures, the system focuses specifically on file path attributes and access patterns, significantly reducing computational requirements while maintaining recommendation quality
Solution Approach 2:
The system uses lightweight file path strings as the basis for recommendations rather than maintaining and processing complex, long-lived user relationship models. File path information is readily available, requires minimal processing, and can be quickly analyzed without the computational overhead of comprehensive user relationship analysis
Data Source
AI summary
Methods, systems and computer program products for recommendation systems. Embodiments commence by gathering a set of pathnames that refer to content objects of a collaboration system. A tokenizer converts at least some of the pathnames into vectors. The vectors comprise hierarchical path components such as folder names or file names, which vectors are labeled with an indication as to whether or not the folder or file referred to in a particular vector had been clicked on by one or more users. Some portion of the labeled vectors are used to train a predictive model. Another portion of the vectors are used to validate the predictive model. When the model exhibits sufficient precision and recall, the predictive model is then used to predict the probability that a particular user would have an interest in a particular folder or file. The folder name or file name is presented as a collaboration recommendation.