Group items tagged

Novelist Kurt Vonnegut once opined, "There is no reason why the simple shapes of stories can't be fed into computers. They are beautiful shapes." Thanks to new text-mining techniques, this has now been done. Researchers at the University of Vermont's Computational Story Lab have analysed over 1,700 English novels to reveal six basic story types - you could call them archetypes - that form the building blocks for more complex stories. They are: 1. Rags to riches - a steady rise from bad to good fortune 2. Riches to rags - a fall from good to bad, a tragedy 3. Icarus - a rise then a fall in fortune 4. Oedipus - a fall, a rise then a fall again 5. Cinderella - rise, fall, rise 6. Man in a hole - fall, rise The researchers used sentiment analysis to get the data - a statistical technique often used by marketeers to analyse social media posts in which each word is allocated a particular 'sentiment score', based on crowdsourced data. Depending on the lexicon chosen, a word can be categorised as positive (happy) or negative (sad), or it can be associated with one or more of eight more subtle emotions, including fear, joy, surprise and anticipation. For example, the word 'happy' is positive, and associated with joy, trust and anticipation. The word 'abolish' is negative and associated with anger. Do sentiment analysis on all the words in a novel, poem or play and plot the results against time, and it's possible to see how the mood changes over the course of the text, revealing a kind of emotional narrative. While not a perfect tool - it looks at words in isolation, ignoring context - it can be surprisingly insightful when applied to larger chunks of text

A traveling journalist in Beijing utilizes both Baidu (China's version of Google) and Google voice-translation apps with mixed results. You speak into the apps, they listen and then translate into the language you choose. They do it in writing, by displaying text on the screen as you talk; and out loud, by using your phone's speaker to narrate what you've said once you're done talking. Typically exchanges are brief: 3-4 turns on average for Google, 7-8 for Baidu's translate app. Both Google and Baidu use machine learning to power their translation technology. While a human linguist could dictate all the rules for going from one language to another, that would be tedious, and yield poor results because a lot of languages aren't structured in parallel form. So instead, both companies have moved to pattern recognition through "neural machine translation." They take a mountain of data - really good translations - and load it into their computers. Algorithms then mine through the data to look for patterns. The end product is translation that's not just phrase-by-phrase, but entire thoughts and sentences at a time. Not surprisingly, sometimes translations are successes, and other times, epic fails. Why? As Macduff Hughes, a Google executive, notes, "there's a lot more to translation than mapping one word to another. The cultural understanding is something that's hard to fully capture just in translation."