Automated API Mashup Generation via Keyword Clustering

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Identifying suitable APIs for application development is burdensome due to the need for manual searches across diverse platforms, limited information availability, and the complexity of verifying API combinations.

Innovation Solution

A method for generating API mashups involves grouping APIs into sub-clusters based on keywords, identifying keyword combinations using real-world data, and determining similarity scores to automatically rank and recommend API mashups, leveraging text mining and natural language processing techniques.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If manual searches are used to identify suitable APIs, then developers can find API information, but the process is burdensome and time-consuming

Engineering Contradiction:

Improvetime required to identify APIsVSAvoidease of identifying APIs

Core Design Contradiction:

Loss of timeVSEase of operation

Solution Approach 1:

The system performs automatic API mashup generation and ranking without requiring manual developer intervention. The processor automatically groups APIs into sub-clusters, identifies keyword combinations, generates possible mashups, and ranks them using similarity scores based on real-world data, eliminating the need for manual searches and verification

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces manual mechanical search processes with automated computational methods. Text mining and natural language processing techniques automatically analyze API documentation and real-world data to generate and rank mashups, substituting human effort with algorithmic processing

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If developers manually verify API combinations, then they can ensure suitability, but the complexity increases significantly

Engineering Contradiction:

Improveaccuracy of API mashup verificationVSAvoidcomplexity of verification process

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system uses similarity scores as feedback mechanisms to automatically evaluate and rank API mashups. By comparing generated mashups against real-world data and calculating similarity metrics, the system provides automated verification that ensures reliability without requiring complex manual validation processes

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent introduces an automated intermediary system that mediates between API specifications and developer needs. The processor acts as an intermediary by automatically analyzing compatibility, generating mashups, and ranking them based on similarity to real-world usage patterns, eliminating the need for direct manual verification

Inventive Principle:

Principle #24Intermediary (Mediator)

3Productivity

If automated methods are used to generate API mashups, then time and effort are reduced, but the system complexity increases

Engineering Contradiction:

Improvespeed of API mashup identificationVSAvoidcomplexity of generation system

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The system segments the complex task of API mashup generation into distinct automated stages: grouping APIs into sub-clusters based on keywords, identifying keyword combinations from real-world data, generating possible mashups, calculating similarity scores, and ranking results. This segmentation manages complexity by breaking down the overall process into manageable automated steps

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent transforms the qualitative task of API compatibility assessment into quantitative parameter-based processing. By converting API documentation and real-world data into structured parameters and calculating numerical similarity scores, the system enables automated high-speed processing while managing complexity through mathematical rather than logical evaluation

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10191786B2Application program interface mashup generation

Publication Date: 2019.01.29 FUJITSU LTD

AI summary

A method of generating application program interface (API) mashups is provided. The method may include grouping a plurality of APIs into a plurality of sub-clusters based on at least one keyword for each of the plurality of APIs. The method may also include identifying at least one keyword combination for the one or more sub-clusters based on real-world data and two or more keywords for the plurality of sub-clusters. Further, the method may include determining one or more possible API mashups including two or more APIs of the plurality of APIs for the at least one keyword combination. In addition, the method may include determining a similarity score for each possible API mashup of the one or more possible API mashups. The method may also include identifying at least one API mashups from the one or more possible mashups based on the similarity score for each possible API mashup.