China
Africa
Explore chapters and articles related to this topic
A Seamless Integration of Human and Automated Scoring
Published in Duanli Yan, André A. Rupp, Peter W. Foltz, Handbook of Automated Scoring, 2020
Kyle Habermehl, Aditya Nagarajan, Scott Dooley
In this section we provide a high-level overview of machine learning for AES; we note that many of these concepts also apply to short-answer item scoring. A good introduction to machine learning can be found in Murphy (2012) while a good introduction to natural language processing (NLP) can be found in Manning and Schütze (1999). Specifically, computer programming is a process of explicitly developing instructions for a computer to perform some task. However, it is extremely difficult and very expensive to develop explicit instructions for a computer to perform the task of scoring / grading essays. Machine learning is a process of providing data to a computer algorithm that can then “learn on its own” to perform a task without explicit instruction in how to do it. Machine learning is what has made it cost-effective to instruct computers in how to score essays.
Technologies
Published in Henry H. Perritt, Eliot O. Sprague, Domesticating Drones, 2016
Henry H. Perritt, Eliot O. Sprague
Increasing automation means that more of the critical aircraft systems are implemented by computer software rather than mechanical structures, assemblies, connections, and movements. Faults in software are far more likely to be due to mistakes in coding logic than to physical failure. In the world of computer programming, fault analysis and mitigation is known as debugging. The more complex the program, the more difficult it is to debug. A fault may manifest itself in the overall failure of the system of which it is a part. A program may cease execution or produce wrong values, but isolating exactly what fault caused it is challenging. Advancing the frontier of reliability engineering for aircraft requires better techniques for fault analysis of computer software and automating them.
Bridge structural safety assessment: a novel solution to uncertainty in the inspection practice
Published in Structure and Infrastructure Engineering, 2023
Francesca Poli, Mattia Francesco Bado, Andrea Verzobio, Daniele Zonta
With the scope of formalizing this set of combination rules, the authors will use a chief unambiguous tool: the computer. In fact, this one enables the execution of mathematical and logical operations on the ground of pre-established commands and processes which, given the same inputs, guarantees every time the extraction of constant, repeatable and bias-free outputs. Then, in order to replace the human mental processes in the inspection practice (𝔹) with a computer program, it is necessary to bridge semantics and the informatic field. A computer/programming language, differently than human speech, can be defined as a series of specific commands and operative directives written as sets of instructions (programs). Whilst in human language words can pick up multiple meanings (even full phrases can acquire unspoken idiomatic meanings) in computer language, the communication is strictly based on specific commands running preestablished actions on the grounds of clear and unique inputs. Semantics-wise, each command has a very specific Meaning and computers associate one meaning to one computer command only.