PDF Solutions has been granted a patent for a robust predictive model. The method involves running multiple predictive models for a target feature, comparing their performance, and selecting the best model based on predefined criteria. A survey can also be used to obtain user preferences for parameters associated with the target feature. The selected model is then deployed in a semiconductor process. GlobalData’s report on PDF Solutions gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
According to GlobalData’s company profile on PDF Solutions, CMOS gate array designing was a key innovation area identified from patents. PDF Solutions's grant share as of September 2023 was 63%. Grant share is based on the ratio of number of grants to total number of patents.
A method for generating a robust predictive model
A recently granted patent (Publication Number: US11775714B2) describes a method for generating a robust predictive model for a target feature in a semiconductor process. The method involves identifying minimum performance criteria for the target feature and generating multiple predictive models to predict the target feature. The models that meet the minimum performance criteria are compared based on true positive and false positive results. The selection of the predictive model is based on defined selection criteria, which can be either manually selected by a user or automatically selected based on user input. The selected predictive model is then deployed into the semiconductor process.
The patent also describes additional steps for the selection process. In one embodiment, the true positive and false positive results are displayed to the user in a user interface, allowing the user to manually select a predictive model based on their preference for a lower number of false positive results with a user-specified constraint on true positive results. In another embodiment, the user can manually select a predictive model based on a preference for a higher number of true positive results with a user-specified constraint on false positive results.
The method further includes the option for automatically selecting a predictive model and setting thresholds for key parameters associated with the target feature based on user utility and/or preferences. The user can provide input through the user interface to establish their preferences and/or utility for the key parameters, and the predictive model is automatically selected based on this input.
To guide the user through the selection process, a survey can be generated and displayed in the user interface. The user's responses to the survey are stored, and their preferences and/or utility for the key parameters are extracted. Mathematical functions can be configured to evaluate and present the trade-offs associated with the key parameters based on the user's preferences and/or utility.
Overall, this patent presents a method for generating a robust predictive model for a target feature in a semiconductor process, allowing for manual or automatic selection based on user preferences and utility. The method provides a user-friendly interface and options for customization to optimize the predictive model selection process.
To know more about GlobalData’s detailed insights on PDF Solutions, buy the report here.
Premium Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.
GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.
GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.
