#DataPlatform #DataAnalysis #AWS #DataScience #DataLake #Jupyter #DataDrivenBusiness

To be strict they are not dead... yet, but they will be if we don’t change our approach.

Firstly, now it takes around 6 months from the start of building a data platform until analytical work becomes possible. Second thing, very often companies don’t know how to monetize data, or they don’t have the resources to use it.

But what if we change our approach and data platform will be built in an analysis-oriented way?

Start by thinking about how data can help your organization,

then do the analysis,

and the platform will be created "by the way" as a response to analytical needs.

Join our presentation and see how we do it at BlueSoft.

#machine_leaning #NLP #DSP #music_industry

The revolution of machine learning is reaching every aspect of our lives - including art and music.
In this talk, we will dive into the world of song analysis and the extraction of lyrical and musical features. We will discuss existing approaches, both in machine learning - Natural Language Processing, Digital Signal Processing - and in music theory & linguistics.

We will see how we can use these features in different kinds of machine learning models, and how these models can be used to solve problems in the music industry, such as song tags and song similarity.
Attend this talk to learn how your technical skills can be useful also to your hobbies.

#recommenderSystems #lean #machineLearning #development

Exploring Lean Data Science for Recommendation Systems

Showcasing the Solution Evolution - from heuristics to complex ML models

How different industries and organizations shape the approach of designing a recommendation systems

#activelearning #machinelearning #deeplearning #wisesampling

Active Learning - a way for collecting labeled data wisely, thus achieving equivalent or better performance levels with fewer data samples and saving time and money.

Background - what type of methods there is? What is the motivation behind using active learning?

Methods - two different methods will be describe, be demonstrating them on a specific real world use case

Results - what are the results of those methods compared to regular random sampling?

Validations - how can we validate the results? How can we be sure that all the components behaves as we want?

#timeseries, #datascience, #forecasting, #preprocessing, #python

Time series data is fundamental for most model businesses — however its handling involves a variety of common issues that can lead to serious consequences when not treated with care.

In this talk, I will introduce the audience to several such common issues, their causes and effective solutions.

This will include: (1) stabilising complex time series dynamics; (2) imputation of time-clustered missing values; (3) reducing impact of noise on forecasting models.

#feedback #loops #recommender #systems

degenerative feedback loops are present not only in recommender systems anymore

when data distribution shifts, model retraining will usually make the situation worse

talk summarizes how to detect and prevent degenerative feedback loops in production environment when applied to a mobile game

#AI #AnomalyDetection #RCA #Network #PredictiveNetworkMaintenance #AI-EmpoweredNetwork

Have you ever wondered how AI is used in the telco network area?

Orange, as data-driven and AI-powered telco company, place Data and AI at the heart of our innovations for Smarter Networks. In that case, network management is supported by Machine Learning models for detecting anomalies based on different types of network data. During this presentation we will zoom on Predictive Network Maintenance based on system logs from network nodes. Our aim is to shorten root cause analysis with Explainable AI to improve network experts daily job.

Chronon is Airbnb's data management platform specifically designed for ML use cases. Previously, ML practitioners at Airbnb spent roughly 60% of their time on collecting and writing transformations for machine learning tasks. Chronon reduces this task from months to days - by making the process declarative. It allows data scientists to easily define features in a simple configuration language. The framework then provides access to point-in-time correct features - for both - offline model training and online inference. In this talk we will explore the problems that arise in industrial feature engineering context and explain how you can use Chronon to solve them.