NeurIPS 2022 allowed researchers and practitioners to share progress and brainstorm new ideas for advancing machine learning and its related fields.
Time series data is widely used in various fields, such as finance, economics, and engineering. One of the key challenges when working with time series data is detecting level shifts. A level shift occurs when the time series’ mean and/or variance changes abruptly. These shifts can significantly impact the analysis and forecasting of the time series and must be detected and handled properly.
SAS' Kirk Swilley and Tom Sabo showcase how you can use perform text analysis on minimal structured narrative data to spot patterns of possible human trafficking.
SAS' Sylvia Kabisa shows you how an online media company might use SAS to offer targeted discounts through personalized pricing.
SAS' Marinela Profi and Sophia Rowland elaborate on IDC including SAS among the leading platform providers for Machine Learning Operations.
Anuja Nagpal and Yonglin Zhu of SAS R&D reveal how, MLPA – without any code and within a given timeframe – finds an effective pipeline for a data set after applying data preprocessing, feature engineering and modeling with hyperparameter tuning.
SAS' Bahar Biller expounds on the idea that stochastic simulations are large-data generation programs for highly complex and dynamic stochastic systems.
Building on a previous post on how the seqmc action can be used to mine frequent sequences, SAS' Amod Ankulkar explores an alternative algorithm.
The ultimate objective of a churn model is preventing churn by making a retention offer. To determine reasonable values for profit and loss information, consider the outcomes and the actions that you would take given knowledge of these outcomes. For example, the marketing department of a telecommunications company wants to offer a discount to people who are no longer on a fixed-term contract. To prevent churn, the company is willing to make an offer in exchange for a one-year contract extension.
The advantage of using SAS PROC KPCA is that you can preprocess your data so that you can classify groups with nonlinear classification boundaries.