Approach to recommending mashups

Patent metadata

Field	Value
Publication number	US-10606658-B2
Application number	US-201615077451-A
Country	US
Kind code	B2
Filing date	Mar 22, 2016
Priority date	Mar 22, 2016
Publication date	Mar 31, 2020
Grant date	Mar 31, 2020

Abstract

A method of recommending Mashups, including identifying, using a processor, pre-existing Mashups implementing application program interfaces (APIs), where each implemented API has one or more attached Tag(s) including API characteristics; extracting the characteristics from the Tags attached to the API(s) implemented in the Mashup, form a set including all of the characteristics from the APIs implemented in Mashup; identifying one or more API(s) not implemented in the Mashups; extracting the characteristics from the Tags attached to the API(s) not implemented in the Mashup to form another set; identifying API characteristics that are an element of one set, but not an element of the other set, forming a third set of these characteristics; modeling a relationship between API(s) using the sets; calculating the probability of one or more API(s) not implemented in a Mashup being used for new Mashup; and presenting the API(s) to a user for new Mashups.

First claim

What is claimed is:

1. A computer implemented method of recommending Mashups, comprising:
identifying, using a processor, one or more pre-existing Mashups that each implement one or more application program interfaces (APIs), where each implemented API provides access to an associated application program, and has one or more attached Tag(s), and each of the Tag(s) includes one or more API characteristics;
extracting each of the one or more API characteristics from the Tags attached to each of the implemented API(s) to form a first set of all the extracted implemented API characteristics;
identifying one or more API(s) not implemented in the identified existing Mashups; extracting each of the characteristics from the Tags attached to each of the unimplemented API(s) to form a second set of all the extracted unimplemented characteristics; identifying one or more implemented API characteristics that is an element of the first set but not an element of the second set;
forming a third set including the one or more API characteristics that is an element of the first set but not an element of the second set;
modeling, using the processor, a relationship between the unimplemented API(s) having unimplemented characteristics and implemented API(s) having implemented characteristics using the first set and the second set to correlate the unimplemented characteristics with the applicability of the implemented API(s) based on the implemented characteristics, which includes determining an association rule from the first set and the second set that connect unimplemented API(s) with pre-existing Mashups;
calculating the probability of one or more unimplemented API(s) being used for a new Mashup;
selecting the one or more unimplemented API(s) to be presented to a user based on a maximum value of confidence;
creating a new Mashup from the one or more unimplemented API(s) by preparing computer code, using the processor, that integrates functions and formats of the one or more unimplemented API(s), wherein the functions and formats of each of the APIs determine the interaction with the associated application program; and
presenting the one or more unimplemented API(s) to a user for use in the new Mashup.

2. The method of claim 1 , which further comprises communicating via the processor, with one or more remote computer systems over a network to identify the one or more pre-existing Mashups and obtaining the implemented characteristics stored as the one or more Tag(s) attached to each implemented API.

3. The method of claim 2 , wherein the one or more remote computer systems is a server and the one or more pre-existing Mashups are stored in the one or more servers.

4. The method of claim 1 , wherein determining the association rule further comprises forming a Mashup-Tag pair to calculate the maximum value of the confidence for a specific pair of one of the pre-existing Mashups and the unimplemented API.

5. The method of claim 1 , wherein determining the association rule does not require a lower limit of confidence value.

6. A mashup recommendation system, comprising:
a processor and memory;
a mashup analyzer that identifies, using the processor, one or more pre-existing Mashups that each implement one or more application program interfaces (APIs) where each implemented API provides access to an associated application program, and has one or more attached Tag(s) and each of the Tag(s) includes one or more API characteristics, and extract each of the one or more API characteristics from the Tags attached to each of the implemented API(s) to form a first set of all the extracted implemented API characteristics;
identifying one or more API(s) not implemented in the identified existing Mashups;
extracting each of the characteristics from the Tags attached to each of the unimplemented API(s) to form a second set of all the extracted unimplemented characteristics;
identifying one or more implemented API characteristics that is an element of the first set but not an element of second set;
forming a third set including the one or more API characteristics that is an element of the first set but not an element of the second set;
an association rule modeler that models, using the processor, a relationship between the unimplemented API(s) having unimplemented characteristics and implemented API(s) having implemented characteristics using the first set and the second set to correlate the unimplemented characteristics with the applicability of the implemented API(s) based on the implemented characteristics which includes determining an association rule from the first set and the second set that connect unimplemented API(s) with pre-existing Mashups; and
an API proposer that calculates the probability of one or more unimplemented API(s) being used for a new Mashup, select the one or more unimplemented API(s) to be presented to a user based on a maximum value of confidence, and create a new Mashup from the one or more unimplemented API(s) by preparing computer code, using the processor, that integrates functions and formats of the one or more unimplemented API(s), wherein the functions and formats of each of the APIs determine the interaction with the associated application program; and
presenting the one or more unimplemented API(s) to a user for use in the new Mashup.

7. The system of claim 6 , which further comprises a network interface that communicates via the processor, with one or more remote computer systems over a network to identify the one or more pre-existing Mashups and obtaining the implemented characteristics stored as the one or more Tag(s) attached to each implemented API.

8. The system of claim 7 , wherein the one or more remote computer systems is a server and the one or more pre-existing Mashups are stored in the one or more servers.

9. The system of claim 6 , wherein determining the association rule further comprises forming a Mashup-Tag pair to calculate the maximum value of the confidence for a specific pair of one of the pre-existing Mashup(s) and the unimplemented API.

10. The system of claim 6 , wherein determining the association rule does not require requires a lower limit of confidence value.

11. A non-transitory computer readable storage medium comprising a computer readable program for a method of recommending Mashups, wherein the computer readable program when executed on a computer causes the computer to perform the steps of:
identifying, using a processor, one or more pre-existing Mashups that each implement one or more application program interfaces (APIs), where each implemented API provides access to an associated application program, and has one or more attached Tag(s), and each of the Tag(s) includes one or more API characteristics;
extracting each of the one or more API characteristics from the Tags attached to each of the implemented API(s) to form a first set of all the extracted implemented API characteristics;
identifying one or more API(s) not implemented in the identified existing Mashups; extracting each of the characteristics from the Tags attached to each of the unimplemented API(s) to form a second set of all the extracted unimplemented characteristics;
identifying one or more implemented API characteristics that is an element of the first set but not an element of the second set;
forming a third set including the one or more API characteristics that is an element of the first set but not an element of the second set;
modeling, using the processor, a relationship between the unimplemented API(s) having unimplemented characteristics and implemented API(s) having implemented characteristics using the first set and the second set to correlate the unimplemented characteristi

Assignees

• IBM

Inventors

• Masuda Satoshi
• Suzuki Akiko
• Nakamura Hiroaki
• Takase Toshiro

Classifications

• G06F9/5072 Primary

  Grid computing · CPC title