AI that can recommend a profession perfect for that person may appear just by analyzing tweets


by

NomadSoul1

`` Tweet Analytics '' that can estimate annual income, age, occupation etc. by Twitter analysis appeared in May 2019, but in the future we recommend more advanced `` profession that is perfect for that person '' from Twitter posts There is a possibility that AI will appear.

Social media-predicted personality traits and values can help match people to their ideal jobs | PNAS
https://www.pnas.org/content/early/2019/12/10/1917942116

Unveiling New Map That Reveals The Hidden Personalities Of Jobs-Eurasia Review
https://www.eurasiareview.com/18122019-unveiling-new-map-that-reveals-the-hidden-personalities-of-jobs/

Robot career advisor: AI may soon be able to analyze your tweets to match you to a job
https://theconversation.com/robot-career-advisor-ai-may-soon-be-able-to-analyse-your-tweets-to-match-you-to-a-job-128777



A research group such as Marian Andrei Rizoiu, who teaches computer science at the Sydney Institute of Technology, analyzed about 128,000 Twitter users who have 3513 jobs, I examined occupational relationships. Specifically, using IBM's artificial intelligence (AI)

Watson , based on the words used in the subject's tweets, create a `` personality prediction digital fingerprint '' and classify it by occupation. did.

The following is a visualization of the analysis results, with 1035 characters in 9 occupations represented by 5 colored dots. Occupation from left: 'Software developer' 'Star scientist' 'First class chemistry researcher' 'Chief information officer' 'Female futurist' 'Architects' 'Doctor' 'Primary school librarian' 'Pro tennis player' 'is. In addition, blue dots represent “openness”, dark orange dots represent “honesty”, gray dots represent “extroverted”, yellow dots represent “friendlyness”, and green dots represent “mental stability”. The higher the position of each dot, the stronger the tendency.



For example, the leftmost software developer and the scientist next to it have many blue dots on the top and yellow dots on the bottom, so you can see that `` it is an open personality but it is unfriendly '' .



In addition, the chief information officer seems to have more green dots at the top, and the mind tends to be stable.



In addition, professional tennis players, in contrast to software developers, have many yellow dots on the top and blue dots on the bottom, so `` friendly, but not very open '' You can also see.



As for the results of the analysis, Mr. Rizoiu said, “Software programmers and scientists are generally open to experiencing various new activities, are intellectually curious, tend to think about symbols and abstractions, and are often bored. On the other hand, tennis players tended to be honest, social and friendly. ”

The research group also used machine learning to output medoids (representative points that can minimize dissimilarities with other points) based on the `` personality prediction digital fingerprint '' and plotted 1227 types of occupations. 'Occupation map' was created. The following is the “profession map”. For example, “concert manager” includes “promoter” and “event manager”, and “software programmer” includes “engineer” and “game developer”. It is. By accessing this link , you can see a detailed version that shows what kind of job each dot is.



Since 1227 professions could be grouped as 20 clusters, the research group said, “Surprisingly, professions could be grouped only based on personality predictions” I pointed out the relevance.

Based on such analysis data, when the research group created a “ data-driven occupational compass” that is an AI that guesses the job that fits the personality, AI estimates the occupation from the personality of the subject with an accuracy of 70% or more. I was able to do it. “This AI was able to show a very similar profession, if wrong, for example, a poet is a fiction writer, not a technologist,” Rizou said.

Margaret Kahn, co-author of the thesis, “Work is at the heart of life, as most of our wake-up time is occupied by work. Many people want a career that suits them. So, by creating a digital fingerprint from information that people have sent online in their private lives, it will be possible to find a profession that suits their personality and support a happy choice for both individuals and society. '' talked.

in Note, Posted by log1l_ks