May 31, 2017 | Written by: Dr. Mohamed Ahmed
Categorized: Citizen Engagement | Events
Share this post:
Deeper Insights from Social Media Using Cognitive Computing
IBM BlueTAS training at GEOINT 2017 | June 6 | 2 – 4 p.m. CDT
Henry B. Gonzalez Convention Center – River Level 007B | San Antonio
What if…unstructured data, such as social media sentiment, could be parsed and analyzed to reveal actionable insights in the same way as structured data that includes, for example, financial, census and customer purchase records?
Many government agencies face tremendous challenges with increasing amounts of news sources and social, geospatial, and imagery data that are collected and need to be analyzed. Analysts must identify, track, and understand patterns, trends, networks and linkages among people, places, and activities.
While humans are good at these activities – analyzing data, formulating hypotheses, and synthesizing the results – our abilities scale poorly. As a result, the exploding volume, variety, and velocity of available information require ever-greater numbers of analysts and accompanying support.
To address some of these pressing needs, IBM developed an innovative cognitive social media analytics framework – BlueTAS – that automatically extracts key information and reveals relationships expressed in Twitter text data, and examined its potential to deliver near real-time intelligence in the form of high-confidence alerts to military customers at both operational and tactical levels.
BlueTAS integrates IBM technologies including entity and relation extraction, language identification, machine translation, sentiment analysis, image tagging and recognition, and network analysis. BlueTAS is largely automated and can be scaled to include a wide variety of existing and emerging data sources across 34 spoken languages. This scalability is important, because new data sources will continue to emerge rapidly, helping to monitor and understand key intelligence questions necessary to anticipate intent and activity in real time.
To help make the “what if” a “wow, we can!” IBM will conduct a training session on its BlueTAS cognitive social media analytics framework at GEOINT 2017 on June 6.
Attendees will be provided a deep dive into IBM BlueTAS and its multiple functions:
- Correlation between live data and established facts;
- Automated extraction of entities such as names, locations, organizations and events and the relationships that might exist between these entities;
- Automated language identification and translation;
- Sentiment analysis;
- Trend and network analysis; and
- Image tagging and recognition.
Training sessions are add-ons and not included in the Symposium Pass registration. To register for my session, visit: GEOINT training registration. We look forward to seeing you on June 6!
You can view some examples of the BlueTAS solution below.
BlueTAS dashboard showing social profile and network analysis
BlueTAS multi-source dashboard
Using BlueTAS you can ingest/analyze social media using a user-defined geo bounding box
Hashtags, mentions, and word cloud
For more information about IBM’s solutions for the US Federal Government, visit ibm.com/federal.
About the author
Dr. Mohamed N. Ahmed is an IBM Distinguished Engineer, chief scientist, Watson and cognitive solutions CTO, US Public Sector. He is leading a team of scientists and software engineers in projects related to natural language processing, multi-lingual information retrieval from unstructured text, machine learning, signal/image processing, and computer vision. Prior to joining IBM in 2010, Dr. Ahmed was a senior researcher at Lexmark International and an adjunct professor at the Electrical and Computer Engineering Department, University of Louisville (Ke.), teaching graduate courses in random fields and estimation theory and machine learning. He also served as an associate professor at the Systems and Biomedical Engineering Dept., Cairo University, Egypt, from 2005-2010. Dr. Ahmed holds 15 U.S. patents, has filed 25 patents, and has published more than 50 peer-reviewed journal and conference papers. He graduated with highest honors from the University of Louisville with a Ph.D. in computer science and engineering. He obtained his B.S. and M.S. in biomedical engineering from Cairo University in 1989 and 1992, respectively.