Please use this identifier to cite or link to this item: https://ah.nccu.edu.tw/handle/140.119/139315


Title: 軟體工作量預估技術於硬體開發時程的追蹤
Tracking of the Hardware Development Schedule based on Software Workload Estimation
Authors: 吳欣純
Wu, Sin-Chun
Contributors: 張宏慶
Jang, Hung-Chin
吳欣純
Wu, Sin-Chun
Keywords: 軟體工作量預估
硬體研發時程追蹤
專案時間管理
機器學習
深度學習
Software Effort Estimation
Hardware Development Effort Tracking
Project Schedule Management
Machine Learning
Deep Learning
Date: 2022
Issue Date: 2022-03-01 18:21:51 (UTC+8)
Abstract: 精準預估開發工作時間一直是專案管理的重點,準確預估能有效掌握資源與控管成本,軟體工作量預估 Software Effort Estimation)早在1960年 L.Farr [5] 和E.A.Nelson [16] 的研究中提出,早期的研究重點著重於建立標準化的估算模型,透過統計迴歸分析或是專家經驗法則預估完成任務的工作時數,後來隨著機器學習與深度學習的發展,透過機器學習與深度學習訓練模型預估工作時間取代原本預估方式。本研究主要提出軟體工作量預估的概念也可延伸應用在電腦製造產業,使用機器學習與深度學習訓練模型預估硬體研發過程中工作任務需要的時間,進而精準掌握產品研發進度與量產上市時程。本文實驗運用語意分析透過自然語言處理(Natural Language Processing, NLP)抽取問題關鍵字當特徵分析,使用機器學習 Machine Learning, ML)的決策樹 Decision Tree)、隨機森林 Random Forest)、XGBoost eXtreme Gradient Boosting)與深度學習 Deep Learning, DL)的RNN模型分析比較精準度、MMRE與PRED(25),實驗發現決策樹
Decision Tree)比其他三個模型顯示較高的準確度。在此研究證明軟體工作量預估技術也可以用在硬體開發過程中的工作追蹤上。
The accuracy of the development effort estimation on the project management is important. If we predict working effort precisely, we could effectively allocated the company's resource and development's cost. Software Effort Estimation (SEE) has been proposed from L.Farr [5] and E.A.Nelson [16] researches since 1960s. Those researches in 1960s are focus on building standard estimation model. Through statistical regression analysis and expert's experience to predict how long to take to finish development tasks. With the rapid development of Machine Learning and Deep Learning, in recent year researches replace statistical regression analysis and expert's experience estimation technology by training Machine Learning and Deep Learning model. In this article, prove Software Effort Estimation technology can not only use on predict software development task but also on use computer's hardware manufacturing development. By using training Machine Learning and Deep Learning's model to predict working effort on hardware research and development tasks to control the product development schedule and launch time precisely. This article experiment uses the Machine Learning's model Decision Tree, Random Forest, XGBoost (eXtreme Gradient Boosting) and Deep Learning RNN model predict working effort and compare the accuracy. And found that Decision Tree model gets the better accuracy and lower error rate than other three models. This article experiment prove Software Effort Estimation can use on hardware development effort tracking.
Reference: [1] L. Breiman, “Random forests”, Machine learning, vol 45, no 1, bll 5–32, 2001.
[2] N. V. Chawla, K. W. Bowyer, L. O. Hall, en W. P. Kegelmeyer, “SMOTE: synthetic minority over-sampling technique”, Journal of artificial intelligence research, vol 16, bll 321–357, 2002.
[3] T. Chen en C. Guestrin, “XGBoost: A Scalable Tree Boosting System”, arXiv [cs.LG], 08-Mrt-2016.
[4] K. Cho et al., “Learning phrase representations using RNN encoder-decoder for statistical machine translation”, arXiv preprint arXiv:1406. 1078, 2014.
[5] L. Farr, H. J. Zagorski, en M. Santa, Factors that Affect the Cost of Computer Programming. Volume Ii. a Quantitative Analysis. 1964.
[6] J. H. Friedman, “Greedy function approximation: A gradient boosting machine”, Ann. Stat., vol 29, no 5, bll 1189–1232, Okt 2001.
[7] Y. Goldberg, “A primer on neural network models for natural language processing”, J. Artif. Intell. Res., vol 57, bll 345–420, Nov 2016.
[8] I. Goodfellow, Y. Bengio, en A. Courville, Deep Learning. London, England: MIT Press, 2016.
[9] T. K. Ho, “Random decision forests”, in Proceedings of 3rd international conference on document analysis and recognition, 1995, vol 1, bll 278–282.
[10] F. K. Jomana Khaseep en R. Assaf, Non-Algorithmic Software Effort Estimation-Challenges and Solutions. .
[11] H. Lee, Speech.ee.ntu.edu.tw. [Online]. Available: https://speech.ee.ntu.edu.tw/~tlkagk/document/Basic.pdf.
[12] Y. Mahmood, N. Kama, A. Azmi, en M. Ali, “Improving estimation accuracy prediction of software development effort: A proposed ensemble model”, in 2020 International Conference on Electrical, Communication, and Computer Engineering (ICECCE), Istanbul, Turkey, 2020.
[13] C. McCormick, “Word2vec tutorial-the skip-gram model”, Apr-2016. [Online]. Available: http://mccormickml. com/2016/04/19/word2vec-tutorial-the-skip-gram-model, 2016.
[14] T. Mikolov, I. Sutskever, K. Chen, G. Corrado, en J. Dean, “Distributed Representations of Words and Phrases and their Compositionality”, arXiv [cs.CL], 16-Okt-2013.
[15] T. Mikolov, K. Chen, G. Corrado, en J. Dean, “Efficient estimation of word representations in vector space”, arXiv [cs.CL], 16-Jan-2013.
[16] E. A. Nelson en M. Santa, Management handbook for the estimation of computer programming costs. 1967.
[17] M. Pandey, R. Litoriya, en P. Pandey, “Validation of existing software effort estimation techniques in context with mobile software applications”, Wirel. Pers. Commun., vol 110, no 4, bll 1659–1677, Feb 2020.
[18] J. R. Quinlan, “Induction of decision trees”, Mach. Learn., vol 1, no 1, bll 81–106, Mrt 1986.
[19] P. Rai, S. Kumar, en D. K. Verma, “Prediction of Software Effort in the Early Stage of Software Development: A Hybrid Model Prévision de l’effort logiciel dans la première phase de développement d’un logiciel : Un modèle hybride”, IEEE Can. J. Electr. Comput. Eng., vol 44, no 3, bll 376–383, 2021.
[20] J. Ramos, “Using tf-idf to determine word relevance in document queries”, Proceedings of the first instructional conference on machine learning, vol 242, bll 29–48, 2003.
[21] S. Raschka, “Model evaluation, model selection, and algorithm selection in machine learning”, arXiv [cs.LG], 13-Nov-2018.
[22] S. Tariq, M. Usman, en A. C. M. Fong, “Selecting best predictors from large software repositories for highly accurate software effort estimation”, J. Softw. (Malden), vol 32, no 10, Okt 2020.
[23] "CS 230 - Recurrent Neural Networks Cheatsheet", Stanford.edu. [Online]. Available: https://stanford.edu/~shervine/teaching/cs-230/cheatsheet-recurrent-neural-networks.
[24] "GitHub - fxsjy/jieba: 结巴中文分词", GitHub, 2021. [Online]. Available: https://github.com/fxsjy/jieba.
[25] "jieba分詞詳解_鴻煊的學習筆記 - MdEditor", Gushiciku.cn, 2021. [Online]. Available: https://www.gushiciku.cn/pl/gCaO/zh-tw.
[26] "Random Forest - Roshan Talimi", Roshan Talimi. [Online]. Available: http://talimi.se/ml/random-forest/.
[27] "tf-idf - 维基百科,自由的百科全書", Zh.wikipedia.org. [Online]. Available: https://zh.wikipedia.org/wiki/Tf-idf.
Description: 碩士
國立政治大學
資訊科學系碩士在職專班
108971006
Source URI: http://thesis.lib.nccu.edu.tw/record/#G0108971006
Data Type: thesis
Appears in Collections:[資訊科學系碩士在職專班] 學位論文

Files in This Item:

File Description SizeFormat
100601.pdf2745KbAdobe PDF0View/Open


All items in 學術集成 are protected by copyright, with all rights reserved.


社群 sharing