当前位置: X-MOL 学术AI EDAM › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Category, process, and recommendation of design in an interactive evolutionary computation interior design experiment: a data-driven study
AI EDAM ( IF 2.1 ) Pub Date : 2020-05-04 , DOI: 10.1017/s0890060420000050
Weixin Huang , Xia Su , Mingbo Wu , Lijing Yang

Design is a complicated and sophisticated process with numerous existing theories trying to describe it. To verify theories and quantitatively describe the design process, design experiment, and data analysis are crucial and inevitable. However, applying data analysis in the design experiment is tricky and design data is not fully utilized in many aspects. To explore the potential of design experiment data, this paper introduces data-driven research based on an interior design experiment, aiming to reveal the category and process of design by conducting data analysis, visualization, and recommendation. We introduce an interactive evolutionary computation (IEC) design experiment that deals with a simplified interior design task and has already been tested on 230 subjects. Using the data gathered during the experiment, we conduct data analysis and visualization involving methods including Holistic color interval and K-means clustering to show categories and processes in design. Additionally, we train a content-based recommendation system with experiment data to capture user preference and make the IEC system more efficient and intelligent. The analysis and visualization show clear design categories and capture an evident trend towards the final design outcome. The application of the recommendation system brings a prominent improvement to the IEC system. This research shows the great potential of the various data-driven methods in design research.

中文翻译:

交互式进化计算室内设计实验中设计的类别、过程和推荐:一项数据驱动的研究

设计是一个复杂而复杂的过程,有许多现有的理论试图描述它。为了验证理论和定量描述设计过程,设计实验和数据分析是至关重要且不可避免的。然而,在设计实验中应用数据分析是很棘手的,设计数据在很多方面都没有得到充分利用。为探索设计实验数据的潜力,本文介绍了基于室内设计实验的数据驱动研究,旨在通过数据分析、可视化和推荐来揭示设计的类别和过程。我们介绍了一个交互式进化计算 (IEC) 设计实验,该实验处理了一个简化的室内设计任务,并且已经在 230 个科目上进行了测试。使用实验期间收集的数据,我们进行数据分析和可视化,涉及包括整体颜色区间和 K-means 聚类在内的方法,以显示设计中的类别和过程。此外,我们使用实验数据训练基于内容的推荐系统,以捕捉用户偏好并使 IEC 系统更加高效和智能。分析和可视化显示了清晰的设计类别,并捕捉到最终设计结果的明显趋势。推荐系统的应用给IEC系统带来了显着的改进。这项研究显示了各种数据驱动方法在设计研究中的巨大潜力。我们用实验数据训练了一个基于内容的推荐系统,以捕捉用户偏好并使 IEC 系统更加高效和智能。分析和可视化显示了清晰的设计类别,并捕捉到最终设计结果的明显趋势。推荐系统的应用给IEC系统带来了显着的改进。这项研究显示了各种数据驱动方法在设计研究中的巨大潜力。我们用实验数据训练了一个基于内容的推荐系统,以捕捉用户偏好并使 IEC 系统更加高效和智能。分析和可视化显示了清晰的设计类别,并捕捉到最终设计结果的明显趋势。推荐系统的应用给IEC系统带来了显着的改进。这项研究显示了各种数据驱动方法在设计研究中的巨大潜力。
更新日期:2020-05-04
down
wechat
bug