Categories
AI

LinkedIn open-sources Greykite, a library for time series forecasting

Join Transform 2021 this July 12-16. Register for the AI event of the year.


LinkedIn today open-sourced Greykite, a Python library for long- and short-term predictive analytics. Greykite’s main algorithm, Silverkite, delivers automated forecasting, which LinkedIn says it uses for resource planning, performance management, optimization, and ecosystem insight generation.

For enterprises using predictive models to forecast consumer behavior, data drift was a major challenge in 2020 due to never-before-seen circumstances related to the pandemic. This being the case, accurate knowledge about the future remains helpful to any business. Automation, which enables reproducibility, may improve accuracy and can be consumed by algorithms downstream to make decisions.

For example, LinkedIn says that Silverkite improved revenue forecasts for 1-day ahead and 7-day ahead, as well as Weekly Active User forecasts for 2-week ahead. Median absolute percent error for revenue and Weekly Active User forecasts grew by more than 50% and 30%, respectively.

Greykite library

Greykite provides time series tools for trends, seasonality, holidays, and more so that users can fit the AI models of their choice. The library provides exploratory plots and templates for tuning, which define regressors based on data characteristics and forecast requirements like hourly short-term forecast and daily long-term forecast. Tuning knobs provided by the templates reduce the search to find a satisfactory forecast. And the Greykite library has flexibility to customize a model template for algorithms, letting users label (and specify whether to ignore or adjust) known anomalies.

Greykite, which provides outlier detection, can also select the optimal model from multiple candidates using past performance data. Instead of tuning each forecast separately, users can define a set of candidate forecast configurations that capture different types of patterns. Lastly, the library provides a summary that can be used to assess the effect of individual data points. For example, Greykite can check the magnitude of a holiday, see how much a changepoint affected the trend, or show how a certain feature might be beneficial to a model.

Greykite Silverkite

With Greykite, a “next 7-day” forecast trained on over 8 years of daily data takes only a few seconds to produce forecasts. LinkedIn says that its whole pipeline, including automatic changepoint detection, cross-validation, backtest, and evaluation, completes in under 45 seconds.

“The Greykite library provides a fast, accurate, and highly customizable algorithm — Silverkite — for forecasting. Greykite also provides intuitive tuning options and diagnostics for model interpretation. It is extensible to multiple algorithms, and facilitates benchmarking them through a single interface,” the LinkedIn research team wrote in a blog post. “We have successfully applied Greykite at LinkedIn for multiple business and infrastructure metrics use cases.”

The Greykite library is available on GitHub and PyPI, and it joins the many other tools LinkedIn has open-sourced to date. They include Iris, for managing website outages; PalDB, a low-key value store for handling side data; Ambry, an object store for media files; GDMix, a framework for training AI personalization models; LiFT, a toolkit to measure AI model fairness; and Dagli, a machine learning library for Java.

VentureBeat

VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact.

Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the subjects of interest to you
  • our newsletters
  • gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
  • networking features, and more

Become a member

Repost: Original Source and Author Link

Categories
AI

Pegasystems adds enterprise AI tools to simplify analysis and forecasting

Join Transform 2021 this July 12-16. Register for the AI event of the year.


Pegasystems today announced it’s adding the ability to apply AI to business processes midstream to enable companies to determine whether an anticipated outcome will occur as expected.

Pegasystems is also adding a feature that lets companies analyze streaming event data on platforms such as the open source Apache Kafka software that is now being widely employed to enable organizations to transfer data in near real time.

Pega Process AI combines machine learning algorithms, event processing, business rules, natural language processing, and predictive analytics with low-code tools to analyze processes in real time, Pegasystems CTO Don Schuerman said. That approach makes it possible for organizations to make adjustments to processes to, for example, ensure a service level agreement (SLA) is met.

As organizations invest in myriad digital business transformation initiatives, many are discovering the batch-oriented legacy applications that typically process data overnight are not well-suited to driving interactions with customers in near real time, Schuerman noted. As a result, organizations are modernizing applications using platforms such as Kafka that enable them to stream data between applications and platforms. The way legacy applications handle data winds up constraining digital processes that need to process data in real time, he added.

“The shift from batch to reactive real-time processes has become table stakes,” he said.

Support for event streaming will play a critical role in enabling organizations to achieve that goal. Rather than having to wait to analyze the data at rest on a cloud platform, for example, it is possible to analyze streaming event data in transit, Schuerman noted.

Build or buy

As organizations look to infuse AI capabilities into business processes, the tensions that always exist between building a capability versus acquiring it will naturally surface. Pegasystems is making a case for an extensible platform based on AI models it creates and curates within the context of the Pega Platform. That capability makes it simpler for organizations to experiment with AI without having to hire data scientists to construct AI models using various open source toolkits. In contrast, it often takes a data science team several months to construct an AI model that may never make it into a production environment.

Schuerman said that rather than simply experimenting with AI capabilities, organizations should generally work backward from a desired business outcome. That approach reduces the chances an organization will wind up investing time and resources into an AI project that never makes it into a production environment.

AI investments

There’s no doubt organizations of all sizes are investing in various forms of AI as part of larger digital business transformation initiatives. The challenge these organizations face is that most data scientists are not especially well-versed in how any given business process should be optimized, which can lead to a lot of trial and error. Providers of platforms such as Pegasystems make it possible for the average business analyst who knows how to work with low-code tools to apply AI to processes they know intimately. It also makes it easier for them to alter those processes, should the AI models need to be updated or start drifting toward a sub-optimal outcome.

As AI becomes more democratized, the processes it can be applied to far exceed the number of data science experts that will be available anytime soon. Out of both necessity and fear, organizations are going to enable business users to at least experiment with AI before rivals apply those same capabilities. Hopefully, guardrails will ensure AI models are properly vetted so they do more good than harm.

VentureBeat

VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact.

Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the subjects of interest to you
  • our newsletters
  • gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
  • networking features, and more

Become a member

Repost: Original Source and Author Link