Semantic Modelling and Machine Learning Analysis of a Billion+ Electronic Components Products to Support Supply Chain Optimization

对十亿种电子元件产品进行语义建模和机器学习分析，支持供应链优化

• 批准号: 522341-2018
• 负责人: Kantarci, Burak
• 金额: $ 1.82万
• 依托单位: University of Ottawa
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2018
• 资助国家: 加拿大
• 起止时间: 2018-01-01 至 2019-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=655061
• 关键词: Semantic; Modelling; Machine; Learning; Analysis

Since many companies in consumer electronics are moving to nearly-optimal supply chain management, processing of digital**content such as electronic component images containing numeric data accompanied with units such as catalogue information**tables and/or charts of electronic components, still require tremendous human effort intervention, potentially leading to significant**costs without any assurance on the accuracy of the digitization. Therefore, innovative solutions are sought to offer software**services that can semantically model the electronic components through machine intelligence by minimizing the human**intervention in the supply chain optimization.**The objective of this project is to develop a semantic model for over a billion electronic components in digital/scanned catalogues,**eliminate the human intervention in the analysis process and optimize supply chain control for these unstructured electronic**component libraries by introducing many hidden perceptron layers in the conventional artificial neural networks, namely by deep**learning.

由于消费电子产品中的许多公司正在转向接近最佳的供应链管理，因此处理数字 ** 内容（例如，包含数字数据的电子元件图像，伴随有例如电子元件的目录信息 ** 表格和/或图表的单元）仍然需要巨大的人力干预，这可能导致显著的 ** 成本，而对数字化的准确性没有任何保证。因此，寻求创新的解决方案来提供软件服务，通过机器智能对电子元件进行语义建模，最大限度地减少供应链优化中的人为干预。该项目的目标是为数字/扫描目录中超过10亿个电子元件开发语义模型，消除分析过程中的人为干预，并通过在传统人工神经网络中引入许多隐藏的感知器层，即通过深度学习，优化这些非结构化电子元件库的供应链控制。