Clusteromvang kiezen

Crystal is operationeel manager bij een maaltijdbezorgdienst in Houston, Texas. Ze heeft breedte- en lengtegraadgegevens van maaltijdverzoeken van de afgelopen maand. Ze wil meerdere nieuwe "bezorghubs" creëren op de plekken waar de vraag het grootst was. Hiervoor heeft ze een machine learning-wetenschapper gevraagd om een clusteringsalgoritme te gebruiken om de bestellingen op basis van hun locatie in groepen te verdelen.

De machine learning-wetenschapper weet niet zeker hoeveel clusters hij moet maken. Bekijk de resultaten van het clusteringsalgoritme met verschillende aantallen clusters.

Wat is het meest geschikte aantal clusters?

Deze oefening maakt deel uit van de cursus

Data Science begrijpen

Bekijk cursus

Interactieve oefening met praktijkervaring

Zet theorie om in actie met een van onze interactieve oefeningen

Begin oefening

Deze oefening maakt deel uit van de cursus

Data Science begrijpen

SkillTag.level.beginnerSkillTag.label

4.8+

6978 reviews

Begin gratis met de cursus

We'll start the course by defining what data science is. We'll cover the data science workflow and how data science is applied to real-world problems. We'll finish the chapter by learning about different roles within the data science field.

Exercise 1: What is data science?Exercise 2: Data science everywhere!Exercise 3: Importance of data science Exercise 4: Applications of data science Exercise 5: Daily dose of data science Exercise 6: The science of smart assistants Exercise 7: Data science roles and tools Exercise 8: Selecting the right candidate Exercise 9: A data scientist role

Now that we understand the data science workflow, we'll dive deeper into the first step: data collection and storage. We'll learn about the different data sources you can draw from, what that data looks like, how to store the data once it's collected, and how a data pipeline can automate the process.

Exercise 1: Data sources Exercise 2: Sorting data sources Exercise 3: Asthma frequencies Exercise 4: Data types Exercise 5: Classifying data types Exercise 6: Net promoter score Exercise 7: Activity tracker Exercise 8: Data storage and retrieval Exercise 9: Cloud platforms Exercise 10: Querying a database Exercise 11: Which type of database?Exercise 12: Data Pipelines Exercise 13: Data pipeline characteristics Exercise 14: Extract Transform Load

Data preparation is fundamental: data scientists spend 80% of their time cleaning and manipulating data, and only 20% of their time actually analyzing it. This chapter will show you how to diagnose problems in your data, deal with missing values and outliers. You will then learn about visualization, another essential tool to both explore your data and convey your findings.

Exercise 1: Data preparation Exercise 2: The truth is out there Exercise 3: Are you prepared?Exercise 4: Exploratory Data Analysis Exercise 5: Numerical EDA Exercise 6: Visual EDA Exercise 7: Visualization Exercise 8: Interactive dashboards Exercise 9: Improving a dashboard

In this final chapter, we'll discuss experimentation and prediction! Beginning with experiments, we'll cover A/B testing, and move on to time series forecasting where we'll learn about predicting future events. Finally, we'll end with machine learning, looking at supervised learning, and clustering.

Exercise 1: A/B-testen Exercise 2: Een A/B-testworkflow opzetten Exercise 3: Statistische significantie Exercise 4: Tussentijdse resultaten Exercise 5: Voorspellen met tijdreeksen Exercise 6: Tijdreeksgegevens classificeren Exercise 7: Interpreteer een tijdreeksplot Exercise 8: Supervised machine learning Exercise 9: Wanneer gebruik je supervised learning Exercise 10: Features en labels Exercise 11: Modelbeoordeling Exercise 12: Clustering Exercise 13: Supervised vs. Unsupervised Exercise 14: Clusteromvang kiezen

Huidige oefening

Exercise 15: Gefeliciteerd!