Visualization using d3.js and DataMaps to visualize attack within given network. Application take data from is IBM QRadar and with help of RESTful API sent query into d3.js when it geocodes to lat-ling and place into the map. For the different level of attack points and lines colorized and filtered on front-end.
GeoJSON for target the province borders on representing map,
CSS3 for visual color and general styling,
With existed world map we create layer with GeoJSON encoding China provinces (including Taiwan). Next, we create additional layer with data, associated with country/province from base map layer. Then colorize the chloroplast to get the visual color distribution across the map. Also added labels for basic view and additional hover effect with corresponding province/data on it.
Build a classifier that can find the localization site of a protein in yeast, based on 8 attributes (features).
For performing classification there was constructed a 3-layer artifical neural network (ANN) and specially a feed-forward multilayer perception. We have used a stochastic gradient descent with back-propagation to train our ANN.
Develop Prediction Model for webspam and hyperlink analysis designed and trained (with provided data) to achieve certain prediction goals.
We have built model for Spam\Nonspam prediction for links analysis company. We have use Big Data methods for input data size 70+ Gb. There were a lot of text features, which were preprocessed by using TF-IDF, Word2Vec and Features Selecting methods. The columns with date format were changed to timestamp format, and period of page life was extracted. As result we have the percentage prediction for each class: Nospam, Page Spam, Domain Spam.
Build a simulator for inner use which predicts label of time series data.
We created a console script written on R and Bash for production purposes that validates predictive models in specific iterative customer defined way. It comprises iterative data splitting, teaching model, predicting outcomes and evaluation of the model performance.
Develop a classifier to identify characters in captchas. Image preprocessing, training, classification had to be done in Python using standard libraries like OpenCV, Scikit-Image, Scikit Learn, etc.
We have found wise to use Machine learning, it means to teach the program detect needed letters and numbers and return correct result. We had dataset for training, which contained images with captchas and correct answers. We have used it for training the model. Then we applied the model to test dataset (only images) and have got the string with answers.