How do the hyperparameters affect your attribution model and how do you choose which ones to tune?

In our previous blog, we introduce how we use attention mechanism for marketing attribution to relate different positions of a sequence of touch points. However, the performance of the model can depend on the selection of hyperparameters.

In this blog, we explain how we use Azure Machine Learning to select the best model in order to have an accurate representation of marketing performance.

Let’s start!

What is hyperparameter tuning?

In machine learning, hyperparameter optimization or tuning is the problem of choosing a set of optimal hyperparameters for…

Originally posted to my personal blog available here.

At QueryClick, we use Azure Functions as the runtime for a number of our data collection pipelines. Serverless technology has been a recent adoption here and it’s worked out well for us so far. We’re big fans of Python in the data engineering team and quickly building robust collection services works out well for us.

Until it gets to maintaining these services. The only real split in Azure Functions between development / staging / production (dev/stg/prod) environments is running it locally (development) or publishing the function and running it on Azure infrastructure…

Originally posted to my personal blog here.

Databricks Notebooks are commonplace at QueryClick, they’re the combination of notebooks and PySpark for EDA and simple Spark jobs. Notebooks are controversial to say the least; they encourage bad coding standards and have a non-linear flow to their code.

However, they do have their uses.

The Problem

One of our clients runs Adobe Analytics as one of their analytics tools and they wanted to use the data collected by Adobe in our attribution solution. …

Photo by Jason Coudriet on Unsplash

This post is part of a series where we outline our progress in the creation of our internal analytics platform both technologically and from a cultural perspective.

At face value, self-service analytics seems so simple. Get some data, make it accessible, extract value from it, but it doesn’t always go like that.

So far our adventure into this area has been interesting and what we currently lack in deliverables we make up for in learning.

Data at QueryClick

To help understand our learning progress, it’s important to understand how our department works and where our data comes from.

Our data comes from 3…

Marketing attribution is a way of measuring the value of the campaigns and channels that are reaching your potential customers. The point in time when a customer interacts with a channel is called a touchpoint, and a collection of touch points forms a user journey. Users journeys can be long with many touch points that makes it difficult to understand the true high and low impact of each interaction, which can result in an inaccurate division of credit and a false representation of marketing performance.

To overcome this problem, we introduce an attention mechanism for marketing attribution to relate different…

Data collection at QueryClick is growing. As we build out our platform and move closer to completing the next significant version of the platform we are looking to the future and revisiting the way we do things.

Our goal of being able to provide insight to our users in the shortest possible time frame is a driving force behind us looking at different ways we can manage our data.

Excluding our clickstream data, a lot of our data is “small” and comes in a semi-structured format via APIs. …

Originally posted to my personal blog here.

As part of the Microsoft Partner Hack in November 2020, I decided to use this opportunity to try out a new method of ingesting Fluentd logs.

What is Fluentd?

Fluentd is a log collector which takes a declarative config file containing input (or “source”) and output information. Wikipedia defines it as:

a cross platform open-source data collection software project

The main idea is it allows developers to collect log information and send it to a given endpoint of their wishes without having to worry about the implementation of the log collection service itself.

How do we use it at QueryClick?

At QueryClick, we…

QueryClick Tech Blog

Antics from the QueryClick R&D team

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store