Insight

JANUARY 14, 2020

Choosing your loss function and optimizer Finally, in the last block of code, we must compile the model that we just built. We pass 3 parameters: loss, optimizer , and metrics. The goal here is to optimize the weights such that the loss is minimized. The class with the highest probability is the winner.

Metrics 58

Metrics Optimization Modeling Testing 58

The Unofficial Google Data Science Blog

JUNE 30, 2016

It should be noted that inverse probability weighting is not generally optimal (i.e., We do this by describing the methods in terms of loss functions whose expectation is optimized at the true value of the propensity score. This algorithm is implemented in the SuperLearner R package (Polley & van der Laan, 2014).

Statistics 44

Statistics Optimization Modeling Experimentation 44

Occam's Razor

SEPTEMBER 15, 2014

In blue is how much time we spent in 2010 and in blue the time spent in 2014. was the dramatic shift between 2010 to 2014 to mobile content consumption. They will need two different implementations, it is quite likely that you will end up with two sets of metrics (more people focused for mobile apps, more visit focused for sites).

Metrics 141

Metrics Reporting Analytics Marketing 141

CIO Business Intelligence

OCTOBER 11, 2023

When you look at the customer journey, many companies say they’re customer obsessed, but you bring that into the DNA by looking at the input metrics that impact the customer experience, not the output metric. Because if you start looking at output metrics that impact the customer experience, it’s too late.

Digital Transformation 78

Digital Transformation Metrics Optimization Strategy 78

IBM Big Data Hub

NOVEMBER 13, 2023

Developed internally at Google and released to the public in 2014, Kubernetes has enabled organizations to move away from traditional IT infrastructure and toward the automation of operational tasks tied to the deployment, scaling and managing of containerized applications (or microservices ).

Machine Learning 117

Machine Learning Data-driven Software Testing 117

CIO Business Intelligence

DECEMBER 18, 2023

During its long-term collaboration with SAP, Hoffenheim has achieved many things including assessing the health and performance of players, personalizing fan experiences to increase engagement and loyalty, and migrating and optimizing its business processes to a single platform to support international growth.

Sales 52

Sales Finance Optimization Software 52

AWS Big Data

JANUARY 25, 2023

For example, in the following table, we would load all carrier/city combinations nested under Dec 7, 2014 before we can continue querying the next date. As users scroll vertically or horizontally on the visual, new queries are run to fetch additional rows and columns of data with fixed row and column configurations for every query request.

Slice and Dice 72

Slice and Dice Visualization Metrics Dashboards 72

Sisense

JUNE 9, 2021

And also like their counterparts in the business world, coaches are relying on metrics to guide their decision-making. In training, wearable devices measure players’ workload, movement, and fatigue levels to manage their fitness and positioning and optimize their performance during play.

Analytics 98

Analytics Data-driven Internet of Things Measurement 98

AWS Big Data

JANUARY 22, 2024

Founded in 2014, Acast is the world’s leading independent podcast company, elevating podcast creators and podcast advertisers for the ultimate listening experience. Through the lenses of the tool, Acast was able to address better monitoring, cost optimization , performance, and security.

Data-driven 83

Data-driven Advertising Metadata Data Architecture 83

Sisense

FEBRUARY 24, 2021

To help provide guidance for what role a chief data officer should play at a particular organization, Yang Lee and a research team introduced their cubic framework for the chief data officer in their seminal 2014 paper for MIS Quarterly Executive.

Data Warehouse 69

Data Warehouse Big Data Metrics Visualization 69

datapine

MAY 2, 2019

Data-driven storytelling is a powerful force as it takes stats and metrics and puts them into context through a narrative that everyone inside or outside of the organization can understand. In our example above, we are showing Sales by Payment Method for all of 2014. They display relationships in how data changes over a period of time.

Visualization 173

Visualization Dashboards Sales KPI 173

datapine

FEBRUARY 23, 2023

The metrics are different and useful independently, but together, they tell a compelling story. Also, showing the relationship between selected multiple variables is not optimal for waterfall charts (also known as Cascade charts), as bubble plots or scatter plots would be a more effective solution.

Visualization 177

Visualization Dashboards Sales Measurement 177

AWS Big Data

JULY 25, 2023

As their workload evolved, Alcion engineers tracked OpenSearch domain utilization via the provided Amazon CloudWatch metrics, making changes to increase storage and optimize their compute resources. This allowed Alcion to focus on optimizing the tenancy model for the new search architecture.

Cost-Benefit 71

Cost-Benefit Modeling Management Optimization 71

Domino Data Lab

AUGUST 22, 2019

At the time—in 2014—the three were colleagues working. You can home in on an optimal value by specifying, say, 32 dimensions and varying this value by powers of 2. If we were using CBOW, then a window size of 5 (for a total of 10 context words) could be near the optimal value. Pennington, J., GloVe: Global vectors. 0.85 = 0.15.

Deep Learning 111

Deep Learning Modeling Metrics Testing 111

Occam's Razor

JUNE 11, 2014

In 2014 those numbers are 5:46 for digital and 4:28 on TV. In early 2014 that number is 2 hours and 51 minutes! It is 2014 after all. Then we did the next most optimal thing, focused on creating "bold standard" reviews. Take any metric you want. Pick any metric you want. How crazy is that?

Marketing 150

Marketing Strategy Sales IT 150

Occam's Razor

SEPTEMBER 21, 2015

The Directors, the Marketers, the Optimization employees and our resident social media gurus. You can cut yourself with it and embarrass yourself, or you can look the very best you ever have by using it optimally. First, someone worked really hard on this and created a really nice model for a smarter decision to be made for 2014.

Visualization 160

Visualization Key Performance Indicator Slice and Dice Strategy 160

Occam's Razor

MARCH 10, 2015

The other big reason for Universal Analytics is this, the column in the middle: The data is from the Q4 2014 Quarterly Retail E-Commerce Sales from the US Census Bureau. You can't really take your offline data about me (family person, xx age, loves to buy from your catalog) and optimize my online experience. You can now do both.

Strategy 151

Strategy Analytics Measurement Business Analysis 151

Domino Data Lab

FEBRUARY 4, 2019

Curiosity as an antidote for a typical enterprise anti-pattern: in the absence of narrative, people make up stories (read: myths) from whatever partial data is at hand that serves the immediate interests(see chapter 2 in Understanding Context by Andrew Hinton (2014). Echoing Amy’s point about ownership as a driver for teams.

Data Science 66

Data Science Experimentation Data-driven Machine Learning 66

datapine

JANUARY 31, 2022

This benefit goes directly in hand with the fact that analytics provide businesses with technologies to spot trends and patterns that will lead to the optimization of resources and processes. Lieferando is a European online food-ordering service that was acquired by Just Eat Take Away in 2014. Let’s see it with a real-world example.

Business Intelligence 214

Business Intelligence Analytics Cost-Benefit ROI 214

Occam's Razor

NOVEMBER 18, 2014

And sometimes they are indeed optimal: 7 Data Presentation Tips: Think, Simplify, Calibrate, Visualize. You can hover over each box to get a sense of the key metrics. Better than the table, but perhaps less optimal than the Treemap. We know the optimization that is required. Our lives are dominated by columns and rows.

Visualization 143

Visualization Slice and Dice Predictive Modeling Metrics 143

Domino Data Lab

AUGUST 1, 2021

Because of its architecture, intrinsically explainable ANNs can be optimised not just on its prediction performance, but also on its explainability metric. Joint training, for example, adds an additional “explanation task” to the original problem and trains the system to solve the two “jointly” (see Bahdanau, 2014).

Modeling 139

Modeling Deep Learning Machine Learning Knowledge Discovery 139

Occam's Razor

FEBRUARY 9, 2015

. + Traffic Trends Key Metrics Analysis. Onsite Behavior Key Metrics Analysis. I tend to use Hitwise a lot less, or not at all, because it is a very hard to use, it is slow, the UX/UI, metrics and reports have not really evolved over the years. Traffic Trends Key Metrics Analysis. Visitor/Audience Type Profile Analysis.

Metrics 155

Metrics Reporting IT Advertising 155

Occam's Razor

NOVEMBER 2, 2015

Success Metrics. In my Oct 2011 post, Best Social Media Metrics , I'd created four metrics to quantify this value. I believe the best way to measure success is to measure the above four metrics (actual interaction/action/outcome). It can be a brand metric, say Likelihood to Recommend. It is not that hard.

B2B 167

B2B Metrics Testing Recreation/Entertainment 167

O'Reilly on Data

NOVEMBER 7, 2019

YouTube’s and Facebook’s algorithms for optimizing “engagement” can make any content viral in seconds. The danger is that online advertising is searching for engagement and virality, and it’s much easier to maximize engagement metrics with faked extreme content. Once you’ve created a fake, you can use social media to propagate it.

Advertising 150

Advertising Metadata Technology Software 150

O'Reilly on Data

JANUARY 25, 2021

in 2008 and continuing with Java 8 in 2014, programming languages have added higher-order functions (lambdas) and other “functional” features. We don’t see that in our data, though there are certainly some metrics to say that artificial intelligence has stalled. Starting with Python 3.0 Neither is challenging the dominant players.

Machine Learning 363

Machine Learning Software Testing Technology 363

Domino Data Lab

SEPTEMBER 9, 2019

For example, with those open source licenses we can download their text, parse, then compare similarity metrics among them: In [12]: pairs = [?. ["mit", "asl"],?. ["asl", "bsd"],?. ["bsd", "mit"] ?]? ?for …. One common use case for natural language work is to compare texts. for a, b in pairs:?.

Deep Learning 110

Deep Learning Machine Learning Visualization Data Science 110