山东理工大学学报(自然科学版)
山東理工大學學報(自然科學版)
산동리공대학학보(자연과학판)
Journal of Shandong University of Technology (Natural Science Edition)
2015年
6期
25-31
,共7页
单构件选择%Vague集%构件质量评价%多构件选择%自适应遗传算法
單構件選擇%Vague集%構件質量評價%多構件選擇%自適應遺傳算法
단구건선택%Vague집%구건질량평개%다구건선택%자괄응유전산법
single component selection%Vague set%evaluation of component quality%multiple component selection%adaptive genetic algorithm
为了有效地进行软件构件选择,将软件构件选择分成单构件选择和多构件选择两类问题。针对单构件选择问题,把改进的Vague集引入构件质量评价,通过评价给出构件质量的优劣排序,将多构件选择问题映射为一个多选择背包问题,提出一种混合的自适应遗传算法,该算法在迭代初期采用固定的交叉概率和变异概率,在迭代末期采用改进的自适应遗传算法。通过仿真实验表明,与其他算法比较该算法具有较好的全局寻优能力和执行效率。
為瞭有效地進行軟件構件選擇,將軟件構件選擇分成單構件選擇和多構件選擇兩類問題。針對單構件選擇問題,把改進的Vague集引入構件質量評價,通過評價給齣構件質量的優劣排序,將多構件選擇問題映射為一箇多選擇揹包問題,提齣一種混閤的自適應遺傳算法,該算法在迭代初期採用固定的交扠概率和變異概率,在迭代末期採用改進的自適應遺傳算法。通過倣真實驗錶明,與其他算法比較該算法具有較好的全跼尋優能力和執行效率。
위료유효지진행연건구건선택,장연건구건선택분성단구건선택화다구건선택량류문제。침대단구건선택문제,파개진적Vague집인입구건질량평개,통과평개급출구건질량적우렬배서,장다구건선택문제영사위일개다선택배포문제,제출일충혼합적자괄응유전산법,해산법재질대초기채용고정적교차개솔화변이개솔,재질대말기채용개진적자괄응유전산법。통과방진실험표명,여기타산법비교해산법구유교호적전국심우능력화집행효솔。
In order to select software component effectively ,the software component selection is divided into two kinds of problem ,one is single component selection and the other is multiple component selection .For the single component selection ,improved vague set is introduced into the evaluation of component quality .By using the evaluation method the ranking of the compo‐nent quality is given .Multiple component selection is mapped as the multiple-choice knapsack problem and we propose a hybrid adaptive genetic algorithm .In the algorithm ,crossover proba‐bility and mutation probability are fixed in the early iterations ,w hile in the last stage of iteration we use improved adaptive genetic algorithm .The simulation experiments show that comparisons with other algorithms the algorithm has strong ability of global optimization and high efficiency .