Head Big Data Analytics, Vodafone India
Paul Pallath is Head Big Data Analytics, Vodafone India. With over 20 years of experience in Machine Learning, Paul has several research publications in the field of Machine Learning & Data Mining in International Journals and conferences and has also invented several patentable ideas. He has a Master’s Degree in Computer Applications with Gold Medal, and PhD in Machine Learning, both from Indian Institute of Technology.
The disclosure generally describes computer-implemented methods, software, and systems, including a method for generating executable components. One method includes identifying a user request to create a new function based pre-existing algorithms, the new function to be used in an application used by a user; providing a set of available algorithms from an algorithm library; receiving a selection by a user of an algorithm from the available algorithms; providing a set of available parameters associated with the selected algorithm; receiving an election by the user of one or more parameters from the set of available parameters; generating an executable component in response to receiving the selection of the algorithm and the election of the one or more parameters, the executable component performing the selected algorithm using at least the elected one or more parameters; and storing the executable component for subsequent execution in response to the requested new function.
Automatic discovery of counter-intuitive insights in data analytics involves computing a first set of values based on primary values and secondary values. The primary values include outliers. The computed first set of values is identified as a primary pattern. Compute a second set of values based on the primary values and first level secondary sub-values. The computed second set of values is identified as a secondary pattern. The identified secondary pattern is opposite to the identified primary pattern. The identified primary pattern and the secondary pattern are displayed in a graphical user interface.
A dataset for an event of interest is received. The dataset represents occurrences of events including data corresponding to features. Event frame sizes are determined to generate insights on the dataset. Features from the occurrences of events are extracted corresponding to the determined event frame sizes. The extracted features are represented as feature abbreviations corresponding to a context. The feature abbreviations with high frequency of occurrence are identified. Rules are generated based on the identified feature abbreviations. Weights are associated to the feature abbreviations variably. Here, the association of weights is based on frequency of occurrence of feature abbreviations in the rules. The features corresponding to feature abbreviations are displayed as insights on the dataset. The displayed features correspond to a high probability of occurrence of the event of interest.
Strategy parameters and weights associated with the strategy parameters are received in a predictive analytics application to dynamically rank entities. Raw values associated with the strategy parameters are normalized by applying transformation functions to get normalized values. Based on the normalized values and the weights associated with the strategy parameters, weighted normalized values are computed. Based on the weighted normalized values aggregate scores are computed. The entities based on the computed aggregate score are dynamically ranked. The dynamically ranked entities in descending order of aggregate scores are displayed in a user interface of the predictive analytics application.