Accurately identifying a household demographic profile based on its television viewing pattern is important for content personalisation, targeted advertising, and programme design. By understanding who is watching what and when, broadcasters can tailor content to match viewers’ interests. Although machine learning can predict household attributes, uncertainty is often high due to overlapping viewing patterns across demographic groups, shared device usage, and limited samples. Thus, this paper applies the Conformal Prediction framework to provide an uncertainty measure for machine prediction. We also introduce a new nonconformity score to improve prediction efficiency. Experiments on a large-scale, imbalanced TV dataset show that our method achieves an average prediction set size of 1.18 and an 82.8% singleton rate at 95% confidence level, outperforming conventional nonconformity measures in terms of both reliability and efficiency.