In recent years, IRS Research, Applied Analytics, and Statistics (RAAS) is increasingly creating graph data models to facilitate fast, efficient queries and visualization of complex relationships between people, places, and things. Using the KGAPI tool (designed and built through TDS prime contract), RAAS supports business customers in providing data driven insights to understand taxpayer behavior and make better informed decisions about service and enforcement. The KGAPI tool is built on top of graph technology to explore and understand knowledge graph data.
See Sparking Transformation: How GPUs Busted Through a Once-Impossible Analytics Job
Tylor Data Services, LLC obtains WBENC Women’s Business Enterprise (WBE) certification From Women Presidents’ Educational Organization - DC as Woman Owned Small Business (WOSB).
viXra Archives publishes 1910.0362 WalkRNN: Reading Stories From Property Graphs
Abstract: WalkRNN, the approach described herein, leverages research in learning continuous representations for nodes in networks, layers in features captured in property graph attributes and labels, and uses Deep Learning language modeling via Recurrent Neural Networks to read the grammar of an enriched property graph. We then demonstrate translating this learned graph literacy into actionable knowledge through graph classification tasks.
Building upon RAAS’s Knowledge Graph tools to enable graph analytics and help further the IRS mission. The option year will be dedicated to the exploration of emerging techniques in graph analytics and graph computational models, incorporating applications of GPU and distributed frameworks.
Concurrently, TDS plans Research and Development in Artificial Intelligence (WalkRNN).
Cornell University Archives publishes 1907.11623 Node Alertness-Detecting changes in rapidly evolving graphs - arxiv.org
Abstract: In this article we describe a new approach for detecting changes in rapidly evolving large-scale graphs. The key notion involved is local alertness: nodes monitor change within their neighborhoods at each time step. Here we propose a financial local alertness application for cointegrated stock pairs
Tylor Data Services, LLC, a firm specializing in advanced graph analytics, deep learning and data science (www.tylordata.com) is working with the IRS’s Research, Applied Analytics & Statistics (RAAS) Data Exploration & Testing (DET) division to develop Knowledge Graph tools to enable graph analytics and help further the IRS mission.
TDS’ graph analytics solution builds on a Knowledge Graph environment and provides tools to enable end-users to search for graph patterns and examine data results. The project goals include exploration of emerging techniques in graph analytics and graph computational models, and applications of GPU and distributed frameworks. TDS was awarded an 8(a) sole source contract with one base year and two option years.
FINRA has success tracking the migration of brokers with a pattern of sales practice abuse. This successful investigation shows the potential of applying tribe and social network analysis to finding bad actors as they move from firm to firm. FINRA Sanctions 10 Former Global Arena Representatives as a Result of FINRA Crackdown on Broker Migration
Regulator bars 10 at New York broker-dealer Global Arena Capital Corp. after investigating brokers who migrated from an expelled firm Finra cracks down on ‘cockroach’ brokers