System and method for software reuse

• US 20040172612A1
• Filed: 02/27/2003
• Published: 09/02/2004
• Est. Priority Date: 02/27/2003
• Status: Active Grant

• Current Assignee: Micro Focus LLC (Open Text Corporation)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Kasravi, Kasra, Patel, Bhupendra N.
• Granted Patent: US 8,676,853 B2
• Time in Patent Office: Days
• US Class Current: 717/101
• CPC Class Codes: G06F 8/36 Software reuse

Abstract

A method and system are presented for accessing existing software. The method includes maintaining a library of software profiles, creating a new software profile to be searched against the library, searching the library to locate profiles similar to the new software profile, and identifying existing software associated with such similar profiles. An exemplary system embodiment of the present invention includes a library of software profiles and associated software, a search engine or process, an interface which accepts user formulated desired software profiles and outputs software located in the search. In operation, a search engine finds software profiles similar to a user formulated desired software profile and the interface outputs software associated with such similar software profiles. In preferred exemplary embodiments, searching the library is accomplished via Case-Based Reasoning, where a similarity score between software profiles is a function of “matching” and “non-matching” weights for profile field values, and also accounts for “synonymal matching weights” which are managed by a synonyms table.

Claims

1. A method of identifying existing software for reuse, comprising:

maintaining a library of software profiles for existing software;

creating a profile for desired software;

searching the library to identify one or more similar existing software profiles based on the desired software profile; and

identifying existing software.

2. The method of claim 1, where said existing software includes at least one of file names, actual software code, inline comments, and any supplemental and/or additional documentation.

3. The method of claim 1, where each of the existing and desired software profiles are formulated as cases, and the searching of the library is performed via case-based reasoning.

4. The method of claim 3, where said cases describe features using at least one of attribute-value pairs, attribute-true/false pairs, attribute-type pairs, and attribute-descriptor pairs.

5. The method of claim 1, where the profiles and the profile for the desired software are expressed as semantic vectors, and said searching of the library is effected via calculating a similarity score.

6. The method of claim 1, where said existing and new software profiles are expressed in terms of a defined taxonomy.

7. The method of claim 6, where said taxonomy is flexible.

8. The method of claim 3, where the identified existing software is linked to its profile.

9. The method of claim 3, where each case includes a set of attributes.

10. The method of claim 9, where searching of the library includes assigning a similarity score to each existing case, wherein said similarity score quantifies a similarity between the fields of the desired software case and those of each existing software case.

11. The method of claim 10, where the similarity score assigns a positive matching weight to a feature match and a negative matching weight to a non-match.

12. The method of claim 11, where the value of a given feature need not be identical between an existing case and the new case for the feature to be interpreted as a match.

13. The method of claim 12, where non-identical but synonymous feature values are assigned a synonymal matching weight according to a synonyms table, where the synonymal matching weight is less than or equal to an identical matching weight.

14. The method of claim 10, where when a desired software case has a value for a given feature, and an existing case being compared to it has no value for that feature, at least one of the following occurs:

the feature is ignored in the calculation of a similarity score, having no effect on said score;

or

a defined absence weight is assigned, where said absence weight ranges form zero to some negative number.

15. A system for accessing software, comprising:

a library of software profiles and associated software;

a search engine or process;

an interface which accepts user formulated desired software profiles and outputs software associated with search engine results.

16. The system of claim 15, where said associated software and outputted software each includes at least one of file names, actual software code, inline comments, and any supplemental and/or additional documentation.

17. The system of claim 15, where said software profiles and desired software profile are expressed as cases, and said searching of the library is effected via case-based reasoning.

18. The system of claim 15, where the system assigns a positive matching weight when a given field'"'"'s values match and a negative matching weight when they do not match.

19. The system of claim 18, wherein the value of a given field need not be identical between an existing case and the new case for the field to be interpreted as a match.

20. The system of claim 19, wherein non-identical but synonymous field values are assigned a synonymal match weighting according to a synonyms table, and wherein the synonymal matching weight is less than or equal to an identical matching weight.

21. The system of claim 15 further comprising a user interface, wherein a user may do at least one of formulate a desired software profile, effect the searching and retrieval of software components using said desired software profile, select a weighting vector profile to be used for a search, select a synonyms table, view search results, receive software components, and input user feedback.

22. The system of claim 15, where said search engine includes an attribute weight manager, which stores weights for at least one of matching, non-matching, synonymal, and absence occurrences used in generating a similarity score between an existing software profile and a desired software profile.

23. The system of claim 22, where said attribute weight manager stores one or more weighting vectors, each corresponding to a different search profile.

24. A computer program product comprising a computer usable medium having computer readable program code means embodied therein, the computer readable program code means in said computer program product comprising means for causing a computer to:

maintain a library of existing software profiles and linked software;

accept a desired software profile;

search the library to locate existing software profiles similar to said desired software profile; and

output software linked to said similar existing software profiles.

25. The computer program product of claim 24, where said existing software and outputted software each includes at least one of file names, actual software code, inline comments, and any supplemental and/or additional documentation.