API呼び出しとメソッド周辺の識別子の実績に基づいたAPI集合推薦手法  [in Japanese] Recommending API Sets for a New Method Based on Surrounding Identifiers and API Usage History  [in Japanese]

Access this Article

Search this Article

Author(s)

Abstract

現代の多くのソフトウェア開発では,膨大なライブラリから必要なAPIを選び出し,それをどのように組み合わせればよいかを調べなければならない.本研究では,開発者が新規作成したいメソッドの名前を記述したときに,そのメソッド本体で使用される可能性のあるAPIの集合を推薦することで,APIの選択を支援する手法を提案する.使用される可能性のあるAPIの推薦には大規模なソースコード集合から抽出した相関ルールを使用する.相関ルールには,メソッド名などの識別子から得た情報と,使用されたAPIとの関係が記録されている.本手法によって開発者に有用なAPI集合を推薦できるかを調査したところ,推薦の上位に適切なAPIが含まれていることを確認した一方,手法に改善の余地があることも分かった.In modern software development, developers have to select and combine appropriate APIs from vast amount of software libraries to implement features. This paper proposes an approach that takes as input a method name which a developer is attempting to create, and suggests APIs that are likely used as a template of method body. By using the template as a reference and/or editing the template, the developer can write the method body. Our approach generates templates from association rules that associate APIs with identifiers such as method names, class names, and field names included in a large set of source files.

In modern software development, developers have to select and combine appropriate APIs from vast amount of software libraries to implement features. This paper proposes an approach that takes as input a method name which a developer is attempting to create, and suggests APIs that are likely used as a template of method body. By using the template as a reference and/or editing the template, the developer can write the method body. Our approach generates templates from association rules that associate APIs with identifiers such as method names, class names, and field names included in a large set of source files.

Journal

  • IPSJ Journal

    IPSJ Journal 56(2), 692-700, 2015-02-15

    Information Processing Society of Japan (IPSJ)

Codes

  • NII Article ID (NAID)
    110009877383
  • NII NACSIS-CAT ID (NCID)
    AN00116647
  • Text Lang
    JPN
  • Article Type
    Journal Article
  • ISSN
    1882-7764
  • Data Source
    NII-ELS  IPSJ 
Page Top