Statistical topic modeling is widely used in political science to study text. Researchers examine documents of varying lengths, from tweets to speeches. There is ongoing debate on how document length affects the interpretability of topic models. We investigate the effects of aggregating short documents into larger ones based on natural units that partition the corpus. In our study, we analyze one million tweets by U.S. state legislators from April 2016 to September 2020. We find that for documents aggregated at the account level, topics are more associated with individual states than when using individual tweets. This finding is replicated with Wikipedia pages aggregated by birth cities, showing how document definitions can impact topic modeling results.

Jacob Keller, a hospital security guard in Bowling Green, Ohio, starts patrolling corridors at midnight. It's quiet, and he spends most of his time alone. The 45-year-old has lost touch with most of his friends, and his wife and kids are usually asleep when he's working.

The technology is geared toward vehicles that specialize in last-mile delivery, which refers to the final step in getting packages from a distribution center to the customer. The market for last-mile delivery has picked up as online shopping has soared during the coronavirus pandemic. Goodyear's new technology, announced Wednesday, is called SightLine and includes a sensor and proprietary machine-learning algorithms that can predict flat tires or other issues days ahead of time, by measuring tire wear, pressure, road-surface conditions and many other factors. The Morning Download delivers daily insights and news on business technology from the CIO Journal team. The surge of last-mile deliveries during the pandemic means that a lot of vehicles are on the road, "stopping and going, hitting curbs, causing damage to the tires, causing breakdowns and congestion," said Richard Kramer, chief executive of Akron, Ohio-based Goodyear.

Cuyahoga Falls, Ohio, May 13, 2021 (GLOBE NEWSWIRE) -- Bespoke AI-driven predictive models have traditionally been expensive and normally have taken months to create and deploy, making them the exclusive domain of only the largest financial institutions. Community banks and credit unions have been at a major competitive disadvantage in using predictive analytics and AI to drive business decisions. Segmint, the global leader in transaction cleansing and analytics for financial institutions, announced the launch of its industry leading AI Platform, a cloud-based and always-on predictive modeling engine which can build and deploy custom predictive models for any financial institution within two weeks. Segmint's AI Platform delivers predictive models at incredible speed and scale, using insights that describe the full universe of customer data. The AI Platform boasts seamless data integration with multiple cores where data flows into the models on a daily basis, continuously updating the insights.

AKRON, OH--Wondering if it was possibly confusing the man for a different guy with a 10 mm nasal bridge and a right earlobe hanging 0.4 mm lower than his left, a Cognitec FaceVACS-VideoScan Unit #121 facial recognition camera expressed frustration Monday after focusing on a man it knew it had seen before and found itself unable to remember his name. I wanna say Richard?" said the embarrassed surveillance algorithm, which despite having access to 20,000 discrete facts about the individual's personality, buying habits, family, friends, workplace, and medical history, simply couldn't put a name to the 2719 points of data comprising his face. "It's right at the tip of my logic board...C'mon, man, he's lived in Iowa City, he had pneumonia last year, his Social Security number starts with 289. Okay, I saw him outside 7-Eleven, then at Barnes & Noble, and then sitting in traffic at 34th and West, and then for four hours straight in front of his laptop. Damn it, I'm so bad with nonwhite faces."

"Should I tell the story about the killer robots?" "I love telling this story. A group of roughly 50 people listened raptly as Weinberg, president of a think tank called the Bay Area Council Economic Institute, addressed a public-policy luncheon in San Carlos, California, on Thursday afternoon. The event--whose guests of honor included Democratic lawmakers Ro Khanna, who represents Silicon Valley, and Tim Ryan, from the Thirteenth District of Ohio--was meant to foster a lively discussion about connecting Silicon Valley "to the heartland." There would also be a sneak peak of a report showing the economic impact of Bay Area companies on every single congressional district in America. Weinberg began his killer story as the crowd--including the mayor of Akron, Ohio; Silicon Valley municipal leaders; health care executives; business school academics; tech investors; representatives from Oracle and Palantir; and a Teamsters rep--made quick work of the carb-heavy boxed lunches.

John Charles Thomas was born in Akron, Ohio and attended Ellet High School. He graduated from Case Western Reserve University majoring in psychology and minoring in mathematics and drama. He received a Ph.D. in experimental psychology from the University of Michigan. His dissertation compared human performance in a problem solving task to that of an early AI system called "The General Problem Solver." After graduate school, Dr. Thomas managed a research project on the psychology of aging at Harvard Medical School.