Vega IT

How we helped Chirrup build a bird sound recognition system that supports sustainable farming.Chirrup has the solution to biodiversity loss, and it lies in bird sounds. They developed a cloud-hosted bird sound recognition system. Facing multiple challenges while bringing the platform to life  Time-consuming improvements of poor-quality audio filesThe client’s primary goal was to create a highly intuitive application that enables fast and accurate bird recognition from audio clips. They faced numerous difficulties, such as poor-quality audio files that required lots of preprocessing.Collecting and validating data, and ensuring platform efficiencyTo the precision of bird sound predictions, we focused on collecting as much data as possible. One of the client’s requirements was improving the accuracy of predictions. We did that by developing an occurrence mask for the output of the model. The team: the data science and software development firepower  The results: a fast-growing platform, 5X faster than similar modelsThe model we developed exceeded the capabilities of its existing counterparts. It is five times faster than similar bird sound recognition solutions. How we achieved those results:A deep learning networkBuilding a scalable and user-friendly systemWe focused on boosting its accuracy while maintaining its user-friendliness and ease of use. The platform’s machine-learning algorithms were tested and optimized to ensure the highest level of accuracy.The back-end infrastructure was designed with scalability in mind.The full tech stackReact, Material UI, and React-Route NestJS MongoDB AWS infrastructure CloudFront AWS services such as Amazon Simple Queue Service (SQS), Simple Email Service (SES), and Simple Storage Service (S3)SQS A custom-built Convolutional Neural Network (CNN) Read more: https://www.vegaitglobal.com/media-center/business-insights/case-study-chirrup

Mapping healthcare delivery in Africa: building an advanced mapping tool for AkrosAkros is a public health organization that works on improving the delivery of health interventions – like vaccine rollouts and malaria prevention programmes – across southern and western Africa. We worked on Reveal, an open-source mapping app that helps field teams to record and analyze health data in any community, no matter how remote.The challenge: public health mapping in remote areasAkros had developed a mapping app called Reveal which uses satellite imagery to understand where people live, what services they have access to and plan for health campaigns to ensure all people have adequate access.The app had basic mapping functionality, and Akros requested us to improve it. Our goal: make it faster, more interactive, more configurable by the end user, and easier to use. The team: data specialistsWe assembled a team of data specialists with experience in mapping apps – people who love to take large, unwieldy sets of data and turn them into something simple, elegant and useful.Results: an optimized, automated platformWe made the maps much more interactive and intuitive - allowing users to zoom in and out, filter out layers of information, and automatically input data into a central database. Its usefulness extends beyond public health programmes too - Reveal can also be used as a census tool, or to trace water resources.Our first task was to prepare the data for maps. The Reveal maps have many layers of information - starting with geography, but adding information about population, infrastructure, water reserves and more.How we got those results:Importing, storing and accessing dataZoom into the detailFind everything fasterReal-time updates and detailed reportsThe tech we used:ReactJS v17Spring BootJava 11PostgreSQLElasticSearchFor the geo- and metadataKafkaRead more: https://www.vegaitglobal.com/media-center/business-insights/case-study-akros

A Chile-based copper mining group, Antofagasta is the longest continuously listed business on the London Stock Exchange.In collaboration with our partner, Emperor, and with the visual identity they built for Antofagasta, we have created a website and online identity that can help position the company even more firmly on the market. Learn more: https://www.vegait.rs/media-center/blog/antofagasta

Machine learning, business understanding: how we used NLP to optimize Argus media insightsArgus Data Insights uses a mix of technology and human input to monitor print, social media, TV and radio – giving detailed, actionable insights into the industry.The challenge: finding the most relevant insights for clientsThe client wanted to improve the accuracy, relevance and speed of delivery for their insights product. To do that, they wanted a software partner to use the latest data science models to create an advanced solution for the future.Results: An NLP-fuelled engine for detailed reports and rich insightsWe used multiple natural language processing (NLP) models to dramatically improve the quality of the media insights Argus could offer its clients. We focused on four areas to improve:Sentiment analysis Named entity recognitionTarget dominance Language detection Using language-agnostic NLPAt every stage of the project, we worked closely with the team at Argus to iterate towards the best result. Microservices to improve speedWe built every NLP model as an independent microservice with its own API – which makes them much faster to use.Working in German for multilingual solutionsArgus works primarily in German, so for this project we did too. It meant diving deep into the technical language of the space, and helped us work closely with Argus to deliver the best result. Flexible compatibility and increased speedWe provided an API for each NLP model, so it could be used independently by any of the client’s products. In combination with Elasticsearch, this solution significantly reduces the time for generating final results.The tech we used for this project:Standard Python for cleaning and tidying data NLP models from Hugging Face JupyterHub for finetuning Different models for different tasks Other tech:FastAPI library Elasticsearch React DevOps technologiesRead more: https://www.vegait.rs/media-center/business-insights/case-study-argus-data-insights/