If we can't find other life in the universe, machines may be able to do it for us.
A new study out of the UK's Plymouth University drew on artificial neural networks (ANNs), to classify planets into five types.
These would estimate a probability of life in each case, which could be used in future interstellar exploration missions.
ANNs are systems that attempt to replicate the way the human brain learns and are one of the main tools used in machine learning.
They're particularly good at identifying patterns that are too complex for a biological brain to process.
The study team trained their network to bunch planets into five different categories based on whether they are most like present-day Earth, early Earth, Mars, Venus or Saturn's moon Titan.
All five of these objects are rocky bodies known to have atmospheres and are among the most potentially habitable objects in our Solar System.
"We're currently interested in these ANNs for prioritising exploration for a hypothetical, intelligent, interstellar spacecraft scanning an exoplanet system at range," says study leader Christopher Bishop.
Atmospheric observations, known as spectra, of the five solar system bodies are presented as inputs to the network, which is then asked to classify them in terms of the planetary type.
As life is currently known only to exist on Earth, the classification uses a "probability of life" metric which is based on the relatively well-understood atmospheric and orbital properties of the five target types.
Bishop has trained the network with over a hundred different spectral profiles, each with several hundred parameters that contribute to habitability.
So far, the network performs well when presented with a test spectral profile that it hasn't seen before.
"Given the results so far, this method may prove to be extremely useful for categorising different types of exoplanets using results from ground-based and near-Earth observatories," says Dr Angelo Cangelosi, the supervisor of the project.
The technique may also be ideally suited to selecting targets for future observations, given the increase in spectral detail expected from upcoming space missions such Esa's Ariel Space Mission and Nasa's James Webb Space Telescope.
Why we struggle with "g"
Despite seeing it millions of times in pretty much every picture book, every novel, every newspaper and every email message, people are essentially unaware of the more common version of the lower-case print letter g.
Most people don't even know that two forms of the letter - one usually handwritten, the other typeset - exist.
And if they do, they can't write the typeset one we usually see.
They can't even pick the correct version of it out of a line-up.
"We think that if we look at something enough, especially if we have to pay attention to its shape as we do during reading, then we would know what it looks like, but our results suggest that's not always the case" says Michael McCloskey, a cognitive scientist at Johns Hopkins University in the US, and co-author of a new study.
"What we think may be happening here is that we learn the shapes of most letters in part because we have to write them in school.
"Looptail g is something we're never taught to write, so we may not learn its shape as well."
Unlike most letters, g has two lower-case print versions.
There's the opentail one that most everyone uses when writing by hand; it looks like a loop with a fishhook hanging from it.
Then there's the looptail g, which is by far the more common, seen in everyday fonts like Times New Roman and Calibri and, hence, in most printed and typed material.
To test people's awareness of the g they tend to write and the g they tend to read, the researchers conducted a three-part experiment.
First, they wanted to figure out if people knew there were two lowercase print gs.
They asked 38 adults to list letters with two lower-case print varieties
Just two named g.
And only one could write both forms correctly.
"We would say: 'There're two forms of g. Can you write them?' And people would look at us and just stare for a moment because they had no idea," study author Kimberly Wong says.
"Once you really nudged them on, insisting there are two types of g, some would still insist there is no second g."
Next, the researchers asked 16 new participants to silently read a paragraph filled with looptail gs, but to say each word with a g aloud.
Immediately after participants finished, having paid particular attention to each of 14 gs, they were asked to write the g that they just saw 14 times.
Half of them wrote the wrong type, the opentail.
The others attempted to write a looptail version, but only one could.
Finally, the team asked 25 participants to identify the correct looptail g in a multiple-choice test with four options. Just seven succeeded.
"They don't entirely know what this letter looks like, even though they can read it," says study co-author Gali Ellenblum, a graduate student in cognitive science.
"This is not true of letters in general. What's going on here?"
This outlier g seems to demonstrate that our knowledge of letters could suffer when we don't write them.
And as we write less and become more dependent on electronic devices, the researchers wonder about the implications for reading.
"What about children who are just learning to read? Do they have a little bit more trouble with this form of g because they haven't been forced to pay attention to it and write it?" McCloskey says.
"That's something we don't really know. Our findings give us an intriguing way of looking at questions about the importance of writing for reading.
"Here is a naturally occurring situation where, unlike most letters, this is a letter we don't write. We could ask whether children have some reading disadvantage with this form of g."