Get involved in the DSN!
Become a member
Showcase your project
Meet the data scientist
Feedback survey
Data science resources
More information
About the DSN
Email the DSNThe information in these articles is provided 'as-is' and Statistics Canada makes no warranty, either expressed or implied, including but not limited to, warranties of merchantability and fitness for a particular purpose. In no event will Statistics Canada be liable for any direct, special, indirect, consequential or other damages, however caused.
Recent articles
Data to Decisions: Visualizations and ML Modeling of Rental Property Data
Topics covered in this article: Data processing and engineering Computer vision
As per the 2021 census, there were 5-million rental households in Canada, which means roughly one-third of Canadian households are renters. However, much of this rental activity occurs privately, leading to limited and inconsistent data. To bridge this knowledge gap, NorQuest college, acquired processed, analyzed, and visualized rental listings from the stakeholder – Community Data Program, for Ontario.
Continue reading: Data to Decisions: Visualizations and ML Modeling of Rental Property Data
Adopting a high Level MLOps Practice for the Production Applications of Machine Learning in the Canadian Consumer Prices Index
Topics covered in this article: Data processing and engineering Text analysis and generation Ethics and responsible machine learning
Responsible application of Machine Learning (ML) within official statistics requires various processes to make sure that ML is developed in a robust and metric driven fashion, and directly tied to solving processing needs of a specific statistical program. These processes can be operationalized as a framework known as ML Operations or MLOps. Focusing on the use case of the Canadian Consumer Prices Index (CPI), this article provides an overview of how various MLOps processes can be built to ensure that ML models that classify unique products to the categories of the CPI classification system adhere to robust quality assurance, transparency, governance, and provenance best practices, ensuring that model decay is addressed and that price statistics calculated on administrative data are robust. The article also categorizes how MLOps could be implemented by providing an overview of a maturity model and focuses on several key components that are important for price statistics.
Identifying Personal Identifiable Information (PII) in Unstructured Data with Microsoft Presidio
Topics covered in this article: Ethics and responsible machine learning
In today’s digital age, organizations collect and store vast amounts of data about their customers, employees, and partners. This data often contains Personal Identifiable Information (PII). With the growing prevalence of data breaches and cyber attacks, protecting PII has become a critical concern for businesses and government agencies alike. In this article, Statistics Canada will take a detailed look at Microsoft Presidio and how it helps organizations in Canada comply with privacy laws.
Other recent articles
Browse articles by topic
Computer vision
- Comparing Optical Character Recognition Tools for Text-Dense Documents vs. Scene Text
- Computer vision models: seed classification project
- Context modelling with transformers: Food recognition
- Data to Decisions: Visualizations and ML Modeling of Rental Property Data
- Extracting Temporal Trends from Satellite Images
- Greenhouse Detection with Remote Sensing and Machine Learning: Phase One
- Image Segmentation in Medical Imaging
- Indigenous Communities Food Receipts Crowdsourcing with Optical Character Recognition
- Reducing data gaps for training machine learning algorithms using a generalized crowdsourcing application
- Self Supervised Learning in Computer Vision: Image Classification
- Tackling Information Overload: How Global Affairs Canada's "Document Cracker" AI Application Streamlines Crisis Response Efforts
- The Rationale Behind Deep Neural Network Decisions
Data processing and engineering
- A new indicator of weekly aircraft movements
- Adopting a high Level MLOps Practice for the Production Applications of Machine Learning in the Canadian Consumer Prices Index
- An image is worth a thousand words: let your dashboard speak for you!
- Building an All-in-One Web Application for Data Science Using Python: An evaluation of the open-source tool Django
- Creating Compelling Data Visualizations
- Data Engineering in Rust
- Data to Decisions: Visualizations and ML Modeling of Rental Property Data
- Deploying your machine learning project as a service
- Designing a metrics monitoring and alerting system
- Extracting Public Value from Administrative Data: A method to enhance analysis with linked data
- Implementing MLOps with Azure
- Making data visualizations accessible to blind and visually impaired people
- MlFlow Tracking: An efficient way of tracking modeling experiments
- Non-Pharmaceutical Intervention and Reinforcement Learning
- The COVID-19 cloud platform for advanced analytics
- Writing a Satellite Imaging Pipeline, Twice: A Success Story
Predictive analytics
- Forecasting power consumption in remote northern Canadian communities
- From Exploring to Building Accurate Interpretable Machine Learning Models for Decision-Making: Think Simple, not Complex
- Modelling SARS-CoV-2 Dynamics to Forecast PPE Demand
- NRCan's Digital Accelerator: Revolutionizing the way Natural Resources Canada serves Canadians through digital innovation
- Unlocking the power of data synthesis with the starter gide on synthetic data for official statistics
- Use of Machine Learning for Crop Yield Prediction
Text analysis and generation
- A Use Case on Metadata Management
- Adopting a high Level MLOps Practice for the Production Applications of Machine Learning in the Canadian Consumer Prices Index
- Applied Machine Learning for Text Analysis Community of Practice: 2021 in review
- Bias Considerations in Bilingual Natural Language Processing
- Chatting About Chatbots: A review of the Chatbot Workshop
- Document Intelligence: The art of PDF information extraction
- Indigenous Communities Food Receipts Crowdsourcing with Optical Character Recognition
- Official Languages in Natural Language Processing
- Text Classification of Public Service Job Advertisements
- Topic Modelling and Dynamic Topic Modelling: A technical review
- Using data science and cloud-based tools to assess the economic impact of COVID-19
- Version Control with Git for Analytics Professionals
- 2021 Census Comment Classification
Ethics and responsible machine learning
- A Brief Survey of Privacy Preserving Technologies
- Adopting a high Level MLOps Practice for the Production Applications of Machine Learning in the Canadian Consumer Prices Index
- Explainable Machine Learning, Game Theory, and Shapley Values: A technical review
- Identifying Personal Identifiable Information (PII) in Unstructured Data with Microsoft Presidio
- Introduction to Privacy-Enhancing Cryptographic Techniques
- Introduction to Cryptographic Techniques: Trusted Execution Environment
- Introduction to Privacy Enhancing Cryptographic Techniques: Secure Multiparty Computation
- Privacy enhancing technologies: An overview of federated learning
- Privacy preserving technologies part three: Private statistical analysis and private text classification based on homomorpic encryption
- Privacy Preserving Technologies Part Two: Introduction to Homomorphic Encryption
- Protected workloads on public cloud
- Responsible use of automated decision systems in the federal government
- Responsible use of machine learning at Statistics Canada
Other
- Production level code in Data Science
- Celebrating women and girls in science: An interview with Dr. Sevgui Erman
- Co-op student explores the power of Big Data
- Data Science Network Newsletter product feedback survey
- Developing Competency Profiles to Shape Data Science in the Public Service
- Developments in machine learning series: Issue three
- Developments in Machine Learning Series: Issue two
- Developments in Machine Learning Series: Series one
- First Data Science Network Directors' Committee Meeting
- Low Code UI with Plotly Dash
- Ottawa to hold World Statistics Congress in July 2023
- The Data Science Network newsletter turns one!