Presented by Charles Parker

A basic problem in computer science is binary classification, in which an algorithm applies a binary label to data based on the presence or absence of some phenomenon. Problems of this type abound in areas as diverse as computational biology, multimedia indexing, and anomaly detection. Evaluating the performance of a binary labeling algorithm is itself a complex task, often based on a domain-dependent notion of the relative cost of "false positives" versus "false negatives". As these costs are often not available to researchers or engineers, a number of methods are used to provide a cost-independent analysis of performance. In this talk, I will examine a number of these methods both theoretically and experimentally. The presented results suggest a set of best practices for evaluating binary classification algorithms, while questioning whether a cost-independent analysis is even possible.

Presented by Erlend Hamberg.

An important problem in genetics is phylogenetic inference: Coming up with good hypotheses for the evolutionary relationship between species – usually represented as a “family tree”. As the amount of molecular data (e.g. DNA sequences) quickly grows, efficient algorithms become increasingly important to analyze this data. A maximum-likelihood approach with models for nucleotide evolution allows us to use all the sequence data, but is a computationally expensive approach. The number of possible trees also grows rapidly as we include more species. It is therefore necessary to use heuristic search methods to find good hypotheses for the “true” tree. Evolutionary algorithms (EA) is a class of such search/optimization algorithms that has been shown to perform well in other areas where the search space is large and irregular. I will explain my approach and my findings from using an evolutionary algorithm for inferring phylogenies from molecular data.

Join us for Algorithms as Recipes for Computer Programming, presented by Rebekah Golden, a local developer at JanRain.

Algorithms sometimes intimidate but really they are just recipes for computers to follow to do repetitive tasks. This is a beginner class on recognizing when an algorithm would be useful, basic methods for creating one in Python, and a quick look at algorithms that have been written by others for common tasks. Presentation is in a room without computers but if you bring yours you can build a computer recipe or two yourself. Discussion and repetitive task descriptions welcome.

This meeting is open to the public. People of all genders are welcome to attend.

Check back for a rescheduled time.

Sunil Rao will be presenting his past research on SNOTEL data using the time series prediction.

Drought is a serious problem in much of the U.S., with the worst conditions across the southern and western parts of the nation. Much of irrigation and recreation facilities depend on proper forecasting of streamflow. The water supply for irrigation largely comes from rivers and creeks, whose streamflow originates from the springtime melting of winter snow. A water supply forecast is a prediction of streamflow volume that will flow past a point on a stream during a specified season, typically in the spring and summer. One of the primary sources for the data is through NRCS SNOTEL( Snow Telemetry) data (available to public as part of tax dollars at work). In this demo, we showcase one such tool (Timeseries ARIMAX model) to forecast Streamflow volume for Deschutes River Basin, OR and later compare with actual data to see how it performed.

If you would like to join the discussion check us out on Zoom https://zoom.us/j/7891236789.

Do you want to learn and share your passion in a supportive community? Knowledge Mavens is an ethos of sharing, creativity, and inspiration.

Our Meetup provides an opportunity to "Show and Tell" followed by feedback and Q&A. You'll have the opportunity to share with our channels such as Meetup, GitHub, YouTube, and Facebook to connect with more passionate people.

The second half of our session we'll collaborate on new topics. The winner wins an award for the most interesting topic and the opportunity to share in an upcoming session.

In this Meetup, we're going to have fun coding together, so bring your laptop if you can. Our plan is to run through coding examples from https://www.freecodecamp.org/. I'll give you a problem, you can try to solve it, then we'll work on it together. We'll be going through the JavaScript algorithms and data structures section.

You're welcome to join us and code other projects as well.

Be inspired! Knowledge Mavens

In this presentation, we'll learn about the mathematical basis of big-o notation and how it's used to characterize the complexity of various algorithms. The presentation is tailored toward for those who already understand the basics of coding (lists, loops, etc) and are preparing for technical interviews. Presented by PDX Code Guild Instructor Matthew Cooper

Chris Thompson, Machine Learning Engineer, will be joining us to discuss Algorithms and Data Structures.

Instead of being seen as concepts or tools that can make a developer's life easier, new and seasoned programmers often fear and ignore data structures and algorithms. To build high-quality, scalable software and web applications or get a coveted job at a FAANG company, you must know and master these topics.

In this talk, Chris will cover:

-What data structures and algorithms are -Why you should learn them and -Materials he has found helpful in learning these topics

Join us Friday, November 18th at 1 pm Pacific Time (2p MT, 3p CT, 4p ET) in our Google Meet room.

RSVP on Meetup: https://bit.ly/TechTalk1118 RSVP on Eventbrite: https://bit.ly/TechTalkE1118

Can't make it live? We will be posting the recording to our Tech Talks playlist on our YouTube Channel: https://bit.ly/TTA-techtalks

Don't miss this great opportunity to learn and network!