AIVC Event – Winchester Cafe Scientifique : ” Identifying the Language of Text”

Basingstoke IVC science enthusiasts invite you to this extra event on a very topical subject by Alan Ratner. The subject of this evenings talk is “Languages of the World: Identifying the Language of Text”.

Regrettably, the public health situation still means it is likely to be some time before we can resume live meetings. We have decided to try out some online events, using Zoom, a popular and well tested platform for online meetings, https://www.zoom.us/ , which many of you may be familiar with by now.

SUMMARY

This talk is in the field of Computational Linguistics which analyzes the bit patterns of human language. It is at the intersection of Computer Science and Linguistics and is closely related to Information Theory.

There are 3 kinds of language: spoken (sequences of phonemes), written (sequences of graphical characters), and text (sequences of characters encoded for computers). This talk will focus on the languages of the world, especially in the form of text. Say you have some text such as a web page or email in a language you do not recognize. Of the many thousands of languages which one is it?

Identifying the language may allow you to forward it to someone who knows the language or to the appropriate automatic translator. Difficulties include: a) some text will contain few words or may be contain more than one language, b) the ratio of linguistic information to web formatting may be quite small making everything look like English, and c) speed requirements may limit you to extremely simple algorithms (if you wish to perform this task on billions or trillions of web pages or real-time on rapidly streaming text) .

This talk will provide a brief introduction to the world’s scripts (alphabets, etc.) and the languages encoded using those scripts and how languages can be identified.

SPEAKER
Alan graduated from the Massachusetts Insititute of Technolgy and Yale University. Specializing in radio propagation in plasmas, Alan responded to a job ad placed by the US National Security Agency looking for someone to study radio propagation in the ionosphere and worked there for 35 years as a communications engineer, with 7 of those years stationed in England. As analogue communications were replaced by digital communications his interest in the physics and engineering of communications was repaced by the linguistics and computation of communications.

After Alan retired from the government he became Chief Knowledge Engineer at Northrop Grumman Information Systems using parallel computing to make sense of vast data sets including text, images, audio, video, network traffic, network security and financial transactions. He retired to England in 2017.

You will need the Zoom client installed on your computer, tablet or phone. Short video on using Zoom here:

https://www.youtube.com/embed/hIkCmbvAHQQ?rel=0&autoplay=1&cc_load_policy=1

Only one ticket needed per screen. If you are sharing with someone, they don’t need a separate ticket.

We will open the meeting from 7pm so people can settle in, chat etc. There will be a break between talk and Q&A when people can chat etc. It’s not the same as live but closer than I thought

A recording will be available after the event on YouTube.

Our Facebook Group https://www.facebook.com/groups/wincafesci/?source_id=1457090591178602 provides an opportunity to share ideas and keep in touch. You are encouraged to post interesting science links and to comment or discuss.

Details
Date:

July 20

Time:

07:30 pm - 09:00 pm

Click to Register: https://www.meetup.com/Manchester-IVC/events/271714861/
Organizer

Manchester IVC

Website: https://www.meetup.com/Manchester-IVC/