One good thing about being stuck at home during the pandemic is that a person can finally get into the habit of listening to "A Way with Words," a radio show that airs on Friday afternoons on New York's WNYE (91.5 FM; check local listings). The hosts, Martha Barnette and Grant Barrett, are the Click and Clack of word talk. Barnette is a writer who has studied Latin and Greek (her books include "A Garden of Words"), and Barrett is a linguist and lexicographer with an ear for contemporary slang. They make a perfect duo. The show is modelled after "Car Talk," though it is broadcast from San Diego, not Cambridge: the hosts laugh a lot, and when people call in they answer by saying, "You have a way with words," which is always nice to hear.
We've now got mansplaining, manspreading, and manterrupting (see: Donald Trump, in the Presidential debates). A thin, flat pastry made of beaten eggs and flour, prepared by a man once a year, typically on Mother's Day. Two or more pieces of bread ingeniously filled with meat, cheese, etc., consumed by a person who is male. Member of a small group of esteemed male experts gathered to share and exchange expertise for the benefit of an audience of non-males. The last (and only) ingredient added by a man when a woman cooks, which everyone agrees really completes the dish.
Researchers used artificial intelligence to analyze adjectives used to describe men and women in 3.5 million books written in English between 1900 and 2008. They contained around 11 billion words. The results from this large data set confirmed what we already knew. "We are clearly able to see that the words used for women refer much more to their appearances than the words used to describe men. Thus, we have been able to confirm a widespread perception, only now at a statistical level," says computer scientist and assistant professor Isabelle Augenstein of the University of Copenhagen's computer science department.
A key element of any sentiment analysis system is the ability to assign a polarity strength value to words appearing within the documents. In this paper we present a novel approach to polarity strength assignment. The approach is knowledge based in that it uses WordNet to build an adjective graph which is used to measure semantic distance between words of known polarity (reference or seed words) and the target word, which is then used to assign a polarity to the target word. We extend previous work in this area by using a small training data set to learn an optimal predictor of polarity strength and to dampen polarity assigned to non-polar adjectives. We also extend the coverage of previous approaches by exploring additional lexical relations not studied previously. The method has been evaluated on a validation set and shows excellent potential in reducing the assignment of spurious polarity and accurately predicting polarity values for polar adjectives.