Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

KEGG(Kyoto Encyclopedia of Genes and Genomes)〔和文〕 (特集 ゲノム医学の現在と未来--基礎と臨床) -- (データベース)

An unsupervised machine learning model for discovering latent infectious diseases using social media data

Deep embedding clustering based on contractive autoencoder

Artificial intelligence in neurodegenerative diseases: A review of available tools with a focus on machine learning techniques.

Document representation is central to modern natural language processing systems including document clustering. Empirical experiments in recent studies provide strong evidence that unsupervised language models can learn context-aware representations in the given documents and advance several NLP benchmark results. However, existing clustering approaches focus on the dimensionality reduction and do not exploit these informative representations. In this paper, we propose a conceptually simple but experimentally effective clustering framework called Advanced Document Clustering (ADC). In contrast to previous clustering methods, ADC is designed to leverage syntactically and semantically meaningful features through feature-extraction and clustering modules in the framework. We first extract features from pre-trained language models and initialize cluster centroids to spread out uniformly. In the clustering module of ADC, the semantic similarity can be measured using the cosine similarity and centroids update while assigning centroids to a mini-batch input. Also, we utilize cross entropy loss partially, as the self-training scheme can be biased when parameters in the model are inaccurate. As a result, ADC can take advantages of contextualized representations while mitigating the limitations introduced by high-dimensional vectors. In numerous experiments with four datasets, the proposed ADC outperforms other existing approaches. In particular, experiments on categorizing news corpus with fake news demonstrated the effectiveness of our method for contextualized representations.

ADC: Advanced document clustering using contextualized representations

LGscore: A method to identify disease-related genes using biological literature and Google data.

The identification of disease-related genes and disease mechanisms is an important research goal; many studies have approached this problem by analysing genetic networks based on gene expression profiles and interaction datasets. To construct a gene network, correlations or associations among pairs of genes must be obtained. However, when gene expression data are heterogeneous with high levels of noise for samples assigned to the same condition, it is difficult to accurately determine whether a gene pair represents a significant gene-gene interaction (GGI). In order to solve this problem, we proposed a random forest-based method to classify significant GGIs from gene expression data. To train the model, we defined novel feature sets and utilised various high-confidence interactome datasets to deduce the correct answer set from known disease-specific genes. Using Alzheimer's disease data, the proposed method showed remarkable accuracy, and the GGIs established in the analysis can be used to build a meaningful genetic network that can explain the mechanisms underlying Alzheimer's disease.

/pdf/machine-learning-based-identification-of-genetic-4rj2sfb6l0.pdf

Machine learning-based identification of genetic interactions from heterogeneous gene expression profiles.

IMA: Identifying disease-related genes using MeSH terms and association rules.

A symptom is the physical indication of an unstable state or the beginning of diseases. Symptom analysis is an essential factor in the medical area, where it is used for disease diagnosis, drug prescription, and the development of new pharmaceuticals. Commensurate with its importance, symptom analysis has been the subject of various studies in recent years. However, prior literature on this topic has been largely limited to studying symptoms for a specific disease. Our paper attempts to expand and build on previous studies by introducing a network-based symptom analysis. Symptom analysis that can provide a basis for analyzing symptoms related to various diseases. For a universal symptom analysis system, we proposed a network-based symptom analysis. In order to construct a symptom network, we utilized Medical Subject Heading (MeSH) terms and the PubMed search engine which are maintained and developed by the National Center for Biotechnology Information (NCBI) at the National Library of Medicine (NLM). We identified symptom-disease relationships with two measurements, the term frequency-inverse document frequency (TF-IDF) and frequent occurrence of two terms (co-occurrence) from PubMed articles. Symptom-symptom pairs, which is the outline for symptom network, were built up based on symptom-disease relationships. As a result, we constructed a symptom network with 223 nodes and 5313 edges. Evaluations were performed in two ways, compared with two symptom clusters and demonstrated with previous researches. Additionally, proposed method has shown possibility for a guideline of clinical demonstration and a discovery of potential symptoms pair.

Jeongwoo Kim

Papers

ADC: Advanced document clustering using contextualized representations

LGscore: A method to identify disease-related genes using biological literature and Google data.

Machine learning-based identification of genetic interactions from heterogeneous gene expression profiles.

IMA: Identifying disease-related genes using MeSH terms and association rules.

CSnet: Constructing symptom network based on disease-symptom relationships