Data Science Community
In its simplest form, data science involves developing methods of recording, storing, and analyzing... View more
Organizer:
- Organized by
-
-
July Data Science Community: Answers to Unanswered Questions
-
July Data Science Community: Answers to Unanswered Questions
Hello Data Science Community Members,
Thanks so much for joining us during the July Data Science Community Conversation with Jeff Yankley from Consumers Energy discussing Customer Modeling! For those that could not make it to the meeting this month, we missed you and hope you can join us in August. You can always check out what’s coming up in the month ahead on the Events Calendar in UAI Connect.
@Kevin posted a brief summary of July’s Community Conversation and a link to the recording and presentation earlier today. Go check out those resources, also including the links to the session evaluation, the UAI Excellence Awards, and our research study!
I follow-up with you on the July Community Conversation one more time, because there were a couple of unanswered questions during the meeting that Jeff promised to get answers to. Please find the questions and answers copied below. If you have any other questions for Jeff and his team, please reply to this post and they will be happy to answer for you.
Question 1: Do you have specifics as to what method you used for stacking and creating the ensemble? Was it another machine learning approach or selecting the highest score from each or something else?
Answer 1: For stacking and creating the ensemble we averaged all the scores based on three independent methods. All three methods were very close in accuracy for predicting three program buckets.Question 2: What was the dependent variable for the model for output 1 (and for context, has to do with the Financial Health/COVID-19 risk scoring modeling that was done)?
Answer 2: There is no dependent variable as we used unsupervised learning models, which do not have dependent variables. We used unsupervised learning models based on different clustering/scoring approaches, such as K-Means, LTV modeling, and Isolation Forest.Thanks again and I look forwarding to seeing you online next month!
Leslie
——————————
Leslie Cook
Membership & Digital Engagement Manager
Utility Analytics Institute (UAI)
719-203-8650, lcook@endeavorb2b.com
—————————— -
2 Replies
-
Hello @John,
Thanks for reaching out! I hope you had a wonderful vacation. You can find the link to the meeting PPT deck and the recording of the meeting in the first paragraph of the post below, i.e. in this sentence: “Get caught up on the meeting here in the Online Data Science Community and be sure to keep the discussion going on the group forum.”
The full URL is as follows: https://connect.utilityanalytics.com/viewdocument/july-21-meeting-customer-modelin?CommunityKey=68485d2d-d4d9-408f-bc7b-69b2074650c3 (scroll down passed the recording to find a PDF version of the PPT deck).
Thanks again and please let us know if you need anything else.
Leslie
——————————
Leslie Cook
Membership & Digital Engagement Manager
Utility Analytics Institute (UAI)
719-203-8650, lcook@endeavorb2b.com
——————————
——————————————-
Original Message:
Sent: 07-27-2020 09:17
From: John Kinateder
Subject: July Data Science Meeting Now on Demand + What’s NextHi Kevin. I was on vacation last week. Is there a deck you can share from that meeting?
Original Message:
Sent: 7/23/2020 5:39:00 PM
From: Kevin Praet
Subject: July Data Science Meeting Now on Demand + What’s NextHello Data Science Community Members,
Thank you to everyone who took time to join us this past Tuesday for our discussion around Customer Modeling. For those unable to attend, Jeff Yankley (Director of Customer Intelligence and Analytics at Consumers Energy) led the discussion while focusing on two major Consumers Energy use cases; “Residential Demand Response (DR) Recommender Model” and “Predicting customer risks due to effects of COVID-19.” Throughout the meeting, Jeff honed in on these areas discussing key objectives, challenges, and lessons learned. Get caught up on the meeting here in the Online Data Science Community and be sure to keep the discussion going on the group forum.
I know I sent this link after the meeting but for those who have yet to complete, please take a minute and give us your feedback in our community evaluation form below. Your feedback is essential for the growth of the Data Science Community.
https://www.surveymonkey.com/r/2020UAIComm
Next month, August 18th, we will be having a discussion on “Establishing a Career Path for Analysts and Upskilling Your Team ” and have provided a brief description below. More info will be made available closer to the meeting.
Join the Data Science Community to hear success stories from UAI member utilities on how they established career paths for analysts/data scientists and how they have helped their teams improve and learn new skills. The session will include information on career development, people/talent development in analytics, and upskilling your team.
UAI Awards:
(Nomination deadline now extended through August, 17th)
Honor the very best in the utility analytics profession, recognizing the individuals, teams, and organizations that have achieved and maintained the highest standards of excellence through innovation. For more information and nominations, visit our site here!
UAI Research Study:
In an effort to better serve you, UAI is conducting a short research study to understand your information needs. As a token of our appreciation, each participant may enter a drawing to win one of four $100 Visa gift cards. The survey should take about 10 minutes to complete. Thank you in advance!
If you have any questions please let us know otherwise we thank you for your continued involvement and look forward to connecting next month.
Thanks!
——————————
Kevin Praet
Membership Coordinator
Utility Analytics Institute (UAI)
Boulder CO
315-440-3033
—————————— -
Hi Kevin. I was on vacation last week. Is there a deck you can share from that meeting?
——————————————-
Original Message:
Sent: 7/23/2020 5:39:00 PM
From: Kevin Praet
Subject: July Data Science Meeting Now on Demand + What’s NextHello Data Science Community Members,
Thank you to everyone who took time to join us this past Tuesday for our discussion around Customer Modeling. For those unable to attend, Jeff Yankley (Director of Customer Intelligence and Analytics at Consumers Energy) led the discussion while focusing on two major Consumers Energy use cases; “Residential Demand Response (DR) Recommender Model” and “Predicting customer risks due to effects of COVID-19.” Throughout the meeting, Jeff honed in on these areas discussing key objectives, challenges, and lessons learned. Get caught up on the meeting here in the Online Data Science Community and be sure to keep the discussion going on the group forum.
I know I sent this link after the meeting but for those who have yet to complete, please take a minute and give us your feedback in our community evaluation form below. Your feedback is essential for the growth of the Data Science Community.
https://www.surveymonkey.com/r/2020UAIComm
Next month, August 18th, we will be having a discussion on “Establishing a Career Path for Analysts and Upskilling Your Team ” and have provided a brief description below. More info will be made available closer to the meeting.
Join the Data Science Community to hear success stories from UAI member utilities on how they established career paths for analysts/data scientists and how they have helped their teams improve and learn new skills. The session will include information on career development, people/talent development in analytics, and upskilling your team.
UAI Awards:
(Nomination deadline now extended through August, 17th)
Honor the very best in the utility analytics profession, recognizing the individuals, teams, and organizations that have achieved and maintained the highest standards of excellence through innovation. For more information and nominations, visit our site here!
UAI Research Study:
In an effort to better serve you, UAI is conducting a short research study to understand your information needs. As a token of our appreciation, each participant may enter a drawing to win one of four $100 Visa gift cards. The survey should take about 10 minutes to complete. Thank you in advance!
If you have any questions please let us know otherwise we thank you for your continued involvement and look forward to connecting next month.
Thanks!
——————————
Kevin Praet
Membership Coordinator
Utility Analytics Institute (UAI)
Boulder CO
315-440-3033
——————————
Log in to reply.