Technology Category
- Analytics & Modeling - Machine Learning
- Analytics & Modeling - Natural Language Processing (NLP)
Applicable Industries
- Cement
- Education
Applicable Functions
- Product Research & Development
- Warehouse & Inventory Management
Use Cases
- Chatbots
- Virtual Training
Services
- Data Science Services
- Training
About The Customer
Pixability is a data and technology company that enables advertisers to accurately target the right content and audience on YouTube. They use machine learning to automatically identify and categorize YouTube content, helping advertisers maximize their reach with suitable content and optimize ad spend. Pixability's services are crucial for brands looking to maximize their reach on YouTube, a platform where viewers watch over 700 million hours of content daily. By providing granular insights into the suitability of content for brand alignment, Pixability helps advertisers ensure their ads are seen by the right audience, thereby improving the return on their video ad spend.
The Challenge
Pixability, a data and technology company, provides advertisers with the ability to accurately target content and audiences on YouTube. However, with over 700 million hours of YouTube content being watched daily, Pixability faced the challenge of continuously and accurately categorizing billions of videos to ensure ads run on brand-suitable content. Their existing natural language processing (NLP) model for classifying videos was not performing strongly enough. The process of labeling training data for the machine learning solution was slow due to reliance on external data labeling services that required multiple iterations. Collaboration was constrained due to limited time domain experts and data scientists had to solve for ambiguous labels. Additionally, valuable information within titles, descriptions, content, and tags was difficult to normalize.
The Solution
Pixability turned to Snorkel Flow’s Data-centric Foundation Model Development workflow to build an NLP application in less time than it took a third-party data labeling service to label a single dataset. This workflow allowed Pixability to scale up the number of classes they could classify to over 600 while also increasing model accuracy to over 90%. The team used Snorkel Flow’s Foundation Model Warm Start with zero-shot learning to jump-start training data creation. They then used Foundation Model Prompt Builder to develop and refine prompts to correct out-of-the-box FM errors and pull more domain-specific knowledge from various FMs. They created prompts that asked the FM to classify videos based on the description. This programmatic approach to labeling data using knowledge from foundation models generated 500,000 labeled training data points that were used to train a model with 90% accuracy. The team was also able to unlock multi-label NLP capabilities, providing more specific classifications for videos.
Operational Impact
Quantitative Benefit
Case Study missing?
Start adding your own!
Register with your work email and create a new case study profile for your business.
Related Case Studies.
Case Study
System 800xA at Indian Cement Plants
Chettinad Cement recognized that further efficiencies could be achieved in its cement manufacturing process. It looked to investing in comprehensive operational and control technologies to manage and derive productivity and energy efficiency gains from the assets on Line 2, their second plant in India.
Case Study
Revolutionizing Medical Training in India: GSL Smart Lab and the LAP Mentor
The GSL SMART Lab, a collective effort of the GSL College of Medicine and the GSL College of Nursing and Health Science, was facing a challenge in providing superior training to healthcare professionals. As clinical medicine was becoming more focused on patient safety and quality of care, the need for medical simulation to bridge the educational gap between the classroom and the clinical environment was becoming increasingly apparent. Dr. Sandeep Ganni, the director of the GSL SMART Lab, envisioned a world-class surgical and medical training center where physicians and healthcare professionals could learn skills through simulation training. He was looking for different simulators for different specialties to provide both basic and advanced simulation training. For laparoscopic surgery, he was interested in a high fidelity simulator that could provide basic surgical and suturing skills training for international accreditation as well as specific hands-on training in complex laparoscopic procedures for practicing physicians in India.
Case Study
IoT platform Enables Safety Solutions for U.S. School Districts
Designed to alert drivers when schoolchildren are present, especially in low-visibility conditions, school-zone flasher signals are typically updated manually at each school. The switching is based on the school calendar and manually changed when an unexpected early dismissal occurs, as in the case of a weather-event altering the normal schedule. The process to reprogram the flashers requires a significant effort by school district personnel to implement due to the large number of warning flashers installed across an entire school district.
Case Study
Digital Transformation of Atlanta Grout & Tile: An IoT Case Study
Atlanta Grout & Tile, a Tile, Stone & Grout restoration company based in Woodstock, Georgia, was facing challenges with its traditional business model. Despite steady growth over the years, the company was falling behind the web revolution and missing out on the opportunity to tap into a new consumer base. They were using independent software from different vendors for each of their department information and workforce management. This resulted in a lot of manual work on excel and the need to export/import data between different systems. This not only increased overhead costs but also slowed down their response to clients. The company also had to prepare numerous reports manually and lacked access to customer trends for effective business decision-making.
Case Study
Implementing Robotic Surgery Training Simulator for Enhanced Surgical Proficiency
Fundacio Puigvert, a leading European medical center specializing in Urology, Nephrology, and Andrology, faced a significant challenge in training its surgical residents. The institution recognized the need for a more standardized and comprehensive training curriculum, particularly in the area of robotic surgery. The challenge was underscored by two independent studies showing that less than 5% of residents in Italian and German residency programs could perform major or complex procedures by the end of their residency. The institution sought to establish a virtual reality simulation lab that would include endourological, laparoscopic, and robotic platforms. However, they needed a simulator that could replicate both the hardware and software of the robotic Da Vinci console used in the operating room, without being connected to the actual physical console. They also required a system that could provide both basic and advanced simulation training, and a metrics system to assess the proficiency of the trainees before they performed surgical procedures in the operating theater.