Augmenting word2vec with latent Dirichlet allocation within a clinical application
arXiv:1808.03967v11090 citations
Originality Incremental advance
AI Analysis
This work addresses Alzheimer's disease diagnosis for clinical applications, but it is incremental as it builds on existing methods.
The paper tackled the problem of distinguishing Alzheimer's disease from speech transcripts by combining latent Dirichlet allocation and word embeddings, resulting in two hybrid models achieving F-scores above the current state-of-the-art on the DementiaBank dataset.
This paper presents three hybrid models that directly combine latent Dirichlet allocation and word embedding for distinguishing between speakers with and without Alzheimer's disease from transcripts of picture descriptions. Two of our models get F-scores over the current state-of-the-art using automatic methods on the DementiaBank dataset.