Interactive Elicitation of Knowledge on Feature Relevance Improves Predictions in Small Data Sets

12/07/2016
by   Luana Micallef, et al.
0

Providing accurate predictions is challenging for machine learning algorithms when the number of features is larger than the number of samples in the data. Prior knowledge can improve machine learning models by indicating relevant variables and parameter values. Yet, this prior knowledge is often tacit and only available from domain experts. We present a novel approach that uses interactive visualization to elicit the tacit prior knowledge and uses it to improve the accuracy of prediction models. The main component of our approach is a user model that models the domain expert's knowledge of the relevance of different features for a prediction task. In particular, based on the expert's earlier input, the user model guides the selection of the features on which to elicit user's knowledge next. The results of a controlled user study show that the user model significantly improves prior knowledge elicitation and prediction accuracy, when predicting the relative citation counts of scientific documents in a specific domain.

READ FULL TEXT

page 10

page 18

page 19

page 20

page 22

page 23

page 24

research
02/04/2022

Capturing and incorporating expert knowledge into machine learning models for quality prediction in manufacturing

Increasing digitalization enables the use of machine learning methods fo...
research
12/10/2016

Knowledge Elicitation via Sequential Probabilistic Inference for High-Dimensional Prediction

Prediction in a small-sized sample with a large number of covariates, th...
research
10/13/2017

User Modelling for Avoiding Overfitting in Interactive Knowledge Elicitation for Prediction

In human-in-the-loop machine learning, the user provides information bey...
research
08/06/2016

Transferring Knowledge from Text to Predict Disease Onset

In many domains such as medicine, training data is in short supply. In s...
research
02/15/2018

Simulation assisted machine learning

Predicting how a proposed cancer treatment will affect a given tumor can...
research
12/09/2019

Expert-guided Regularization via Distance Metric Learning

High-dimensional prediction is a challenging problem setting for traditi...
research
11/01/2022

Informed Priors for Knowledge Integration in Trajectory Prediction

Informed machine learning methods allow the integration of prior knowled...

Please sign up or login with your details

Forgot password? Click here to reset