Nav: Home

Could your computer please be more polite? Thank you

June 30, 2020

PITTSBURGH--In a tense time when a pandemic rages, politicians wrangle for votes and protesters demand racial justice, a little politeness and courtesy go a long way. Now researchers at Carnegie Mellon University have developed an automated method for making communications more polite.

Specifically, the method takes nonpolite directives or requests -- those that use either impolite or neutral language -- and restructures them or adds words to make them more well-mannered. "Send me the data," for instance, might become "Could you please send me the data?"

The researchers will present their study on politeness transfer at the Association for Computational Linguistics annual meeting, which will be held virtually beginning July 5.

The idea of transferring a style or sentiment from one communication to another -- turning negative statements positive, for instance -- is something language technologists have been doing for some time. Shrimai Prabhumoye, a Ph.D. student in CMU's Language Technologies Institute (LTI), said performing politeness transfer has long been a goal.

"It is extremely relevant for some applications, such as if you want to make your emails or chatbot sound more polite or if you're writing a blog," she said. "But we could never find the right data to perform this task."

She and LTI master's students Aman Madaan, Amrith Setlur and Tanmay Parekh solved that problem by generating a dataset of 1.39 million sentences labeled for politeness, which they used for their experiments.

The source of these sentences might seem surprising. They were derived from emails exchanged by employees of Enron, a Texas-based energy company that, until its demise in 2001, was better known for corporate fraud and corruption than for social niceties. But half a million corporate emails became public as a result of lawsuits surrounding Enron's fraud scandal and subsequently have been used as a dataset for a variety of research projects.

But even with a dataset, the researchers were challenged simply to define politeness.

"It's not just about using words such as 'please' and 'thank you,'" Prabhumoye said. Sometimes, it means making language a bit less direct, so that instead of saying "you should do X," the sentence becomes something like "let us do X."

And politeness varies from one culture to the next. It's common for native North Americans to use "please" in requests to close friends, but in Arab culture it would be considered awkward, if not rude. For their study, the CMU researchers restricted their work to speakers of North American English in a formal setting.

The politeness dataset was analyzed to determine the frequency and distribution of words in the polite and nonpolite sentences. Then the team developed a "tag and generate" pipeline to perform politeness transfers. First, impolite or nonpolite words or phrases are tagged and then a text generator replaces each tagged item. The system takes care not to change the meaning of the sentence.

"It's not just about cleaning up swear words," Prabhumoye said of the process. Initially, the system had a tendency to simply add words to sentences, such as "please" or "sorry." If "Please help me" was considered polite, the system considered "Please please please help me" even more polite.

But over time the scoring system became more realistic and the changes became subtler. First person singular pronouns, such as I, me and mine, were replaced by first person plural pronouns, such as we, us and our. And rather than position "please" at the beginning of the sentence, the system learned to insert it within the sentence: "Could you please send me the file?"

Prabhumoye said the researchers have released their labeled dataset for use by other researchers, hoping to encourage them to further study politeness.
-end-
In addition to the students, the study's co-authors included several professors from the LTI and the Machine Learning Department -- Barnabas Poczos, Graham Neubig, Yiming Yang, Ruslan Salakhutdinov and Alan Black. The Air Force Research Laboratory, Office of Naval Research, National Science Foundation, Apple and NVIDIA supported this research.

Carnegie Mellon University

Related Adds Words Articles:

Authors' 'invisible' words reveal blueprint for storytelling
The ''invisible'' words that shaped Dickens classics also lead audiences through Spielberg dramas.
Study finds hidden emotions in the sound of words
In the midst of the COVID-19 crisis, it's common to feel stress levels rise every time we hear the word ''virus.'' But new Cornell-led research reveals that the sound of the word itself was likely to raise your blood pressure - even before ''corona'' was added to it.
NIH study finds out why some words may be more memorable than others
In a recent study of epilepsy patients and healthy volunteers, National Institutes of Health researchers found that our brains may withdraw some common words, like ''pig,'' ''tank,'' and ''door,'' much more often than others, including ''cat,'' ''street,'' and ''stair.'' By combining memory tests, brain wave recordings, and surveys of billions of words published in books, news articles and internet encyclopedia pages, the researchers not only showed how our brains may recall words but also memories of our past experiences.
Exploring the use of 'stretchable' words in social media
An investigation of Twitter messages reveals new insights and tools for studying how people use stretched words, such as 'duuuuude,' 'heyyyyy,' or 'noooooooo.' Tyler Gray and colleagues at the University of Vermont in Burlington present these findings in the open-access journal PLOS ONE on May 27, 2020.
How the brain separates words from song
The perception of speech and music -- two of the most uniquely human uses of sound -- is enabled by specialized neural systems in different brain hemispheres adapted to respond differently to specific features in the acoustic structure of the song, a new study reports.
Words matter when it comes to apparel for people living with disabilitie
Brands should consider the language they use when marketing products to this group of consumers, according to a new study from the University of Missouri.
Brain patterns can predict speech of words and syllables
Neurons in the brain's motor cortex previously thought of as active mainly during hand and arm movements also light up during speech in a way that is similar to patterns of brain activity linked to these movements, suggest new findings published today in eLife.
'I predict your words': that is how we understand what others say to us
We are at a fun but noisy party: how can we understand the words someone is saying to us despite the background music and voices?
The brain processes words placed on the right side of a screen more quickly
When reading words on a screen, the human brain comprehends words placed on the right side of the screen faster.
Nearly identical representations of spoken, written words in the brain
The brain activity evoked from processing written or heard semantic information is almost identical, according to research in adults published in JNeurosci.
More Adds Words News and Adds Words Current Events

Trending Science News

Current Coronavirus (COVID-19) News

Top Science Podcasts

We have hand picked the top science podcasts of 2020.
Now Playing: TED Radio Hour

Debbie Millman: Designing Our Lives
From prehistoric cave art to today's social media feeds, to design is to be human. This hour, designer Debbie Millman guides us through a world made and remade–and helps us design our own paths.
Now Playing: Science for the People

#574 State of the Heart
This week we focus on heart disease, heart failure, what blood pressure is and why it's bad when it's high. Host Rachelle Saunders talks with physician, clinical researcher, and writer Haider Warraich about his book "State of the Heart: Exploring the History, Science, and Future of Cardiac Disease" and the ails of our hearts.
Now Playing: Radiolab

Insomnia Line
Coronasomnia is a not-so-surprising side-effect of the global pandemic. More and more of us are having trouble falling asleep. We wanted to find a way to get inside that nighttime world, to see why people are awake and what they are thinking about. So what'd Radiolab decide to do?  Open up the phone lines and talk to you. We created an insomnia hotline and on this week's experimental episode, we stayed up all night, taking hundreds of calls, spilling secrets, and at long last, watching the sunrise peek through.   This episode was produced by Lulu Miller with Rachael Cusick, Tracie Hunte, Tobin Low, Sarah Qari, Molly Webster, Pat Walters, Shima Oliaee, and Jonny Moens. Want more Radiolab in your life? Sign up for our newsletter! We share our latest favorites: articles, tv shows, funny Youtube videos, chocolate chip cookie recipes, and more. Support Radiolab by becoming a member today at Radiolab.org/donate.