首页|"Automated Evaluation of Free-Form Answers and Generation of Actionable Feedback to Multidimensional Reasoning Questions" in Patent Application Approval Process (USPTO 20240054909)
"Automated Evaluation of Free-Form Answers and Generation of Actionable Feedback to Multidimensional Reasoning Questions" in Patent Application Approval Process (USPTO 20240054909)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News-A patent application by the inventors Bakken, Sar a (Houston, TX, US); Newstadt, Michelle (Pittsburgh, PA, US); Rosen, Yigal (Broo kline, MA, US); Rushkin, Ilia P. (Cambridge, MA, US), filed on August 15, 2023, was made available online on February 15, 2024, according to news reporting orig inating from Washington, D.C., by NewsRx correspondents. This patent application is assigned to BrainPOP IP LLC (New York, New York, Unit ed States). The following quote was obtained by the news editors from the background informa tion supplied by the inventors: "Students often interact with computerized learn ing platforms by typing text answers into response fields of a graphical user in terface. To evaluate the student's response, various platforms may automatically apply a list of rubrics to the entered text. However, the scores or grades appl ied by computerized learning platforms based on such rubrics can vary significan tly from the grade or score that would be applied manually (e.g., by a teacher). For example, depending upon the ability level of child learners, the responses will often be "low quality" in terms of formatting (e.g., typographical mistakes , erratic use of punctuation, unorthodox syntaxis, misuse of words having relate d meanings, etc.). Additionally, rubrics typically estimate use of language (e.g ., correctness and sufficient complexity of grammar or vocabulary), while approp riate focus for grading or scoring answers related to various subjects instead r equire that the responses relate to the appropriate subject and present the corr ect logical construction.
BrainPOP IP LLCBusinessCyborgsEmer ging TechnologiesMachine LearningMathematics
NETL
NSTL
