Implementation of advanced algorithms and software development
ADINIS is actively exploring application of advanced machine learning algorithms in modeling, analysis and prediction of large scale multivariate data emerging in experimental biology, chemistry, medicine and other scientific disciplines. We strive to provide our customers with solutions of highest quality that require expert knowledge of computational domains (machine learning, statistical modeling, numerical analysis, and software engineering) as well as good understanding of modeled domain (proteomics, genomics, cytology,…).
Our researchers are in close contact with customers from an early stage of each project, often taking leading roles in analyzing customer’s needs and suggesting and designing appropriate solutions. Typically multiple computational approaches are explored in early prototypes and only the approach best fitting the customer’s needs is then implemented to a final software product. The table below specifies some of the techniques that we have used in our projects or have strong expertise in:
Machine Learning |
---|
SVM |
Logistic regression |
Neural Networks |
PCA, MDS, Factor Analysis |
Decision Trees, Random Forests |
Maximum likelihood, graphical models, EM, probabilistic inference, MCMC |
Cluster analysis, Kmeans, GMM |
Statistics |
---|
Hypothesis testing |
ANOVA |
General Linear Model |
Outlier detection |
Bioinformatics : |
Dynamic programming, Needleman-Wunsch, Smith-Waterman |
Quntitative Trait Locus |
Services
- Processing of customer's data using advanced methods of mathematics and machine learning, modeling of biological systems, consulting.
- Software development according to specific requirements of customers.
- Health application models Development and implementation of prediction and statistical models for health applications.
- Insurance fraud models Advanced statistical models for detection of health insurance frauds.