Models
- What is a model?
- Why should I use decision tree models?
- What kind of algorithm does BigML use to build decision tree models and how does it work?
- Why does one dataset take more time to build a model than another?
- How can I interpret the BigML models and interact with them?
- Can I capture an image of an entire model with all its branches?
Ensembles
- What is an ensemble?
- Why is an ensemble more effective than a single decision tree model?
- How many models should I choose to build a robust ensemble?
- What is the difference between Bagging, Random Decision Forest and Boosted Tree? Which one should I use?
- Is it possible to put more or less weight on each instance when building an ensemble?
- How can I change the instances weight when training a model or an ensemble?
Clusters
- What is a cluster?
- When should I use clusters?
- Which algorithm should I use, K-means or G-means?
- Which types of fields does BigML support to build clusters?
- How does BigML calculate centroids for clusters?
- How does BigML calculate the distance between instances to create clusters?
Anomalies
- What is an anomaly?
- Which algorithm does BigML use for Anomaly Detection?
- How does BigML detect anomalies in my dataset?
- Which types of fields does BigML support to compute the anomaly score?
- Do I need to specify any labels in my data to detect anomalies?
- Can I have missing values to calculate anomalies?
Associations
- What is an association?
- Which algorithm does BigML use for Association Discovery?
- Is there any limitation on fields and instances when creating associations?
- Which types of fields does BigML support to find associations?
- How can I interpret an Association Discovery model?
- How do I know which rules are the most relevant ones? Are there any metrics to summarize all results?