Contact Us

Insights

Subscribe to Our Blog
Menu
Request A Demo
Contact Us
Subscribe to Our Blog
FINALnewwhitelogoNoR.png

Machine Learning
Competition 2019

Cash & Prizes

As a result, the team or individuals with the best solution will win. There will be over $2,500 in cash and prizes. 

Schedule

We'll start the competition during winter break, when the first semester ends, and run through the middle or end of February. Stayed tuned for details.

Teams or Individual

Students are encouraged to team up to solve the anaphora challenge. Teams of 2-4 students are likely optimal, but you can also participate as an individual.

Participants

Open to any Iowa State University undergrad and grad student.  

Machine Learning Competition

Step 1: Problem

The competition will focus on a solution for anaphora in linguistics. Anaphora represents a challenge for natural language processing (NLP) and named entity recognition (NER) concepts in computational linguistics, and since Kingland’s Cognitive/AI solutions are centered around successful usage of NLP and NER within the field of Text Analytics, having a solution for anaphora is important.

Step 2: The Solution Concept

Entrants to the competition are encouraged to use open source NLP and NER technology paired with open source and/or academia-based corpus sets to help with language processing and model training. It is expected that good design and custom coding beyond the use of this open source technology will be required to provide a solution. The English language will be the focus for the competition.

Step 3: Winners

Kingland’s Cognitive/AI engineers will judge the solutions based on elegance of design and their effectiveness in resolving the anaphora challenge on a number of different text/documentation examples. In an absence of a complete or working solution, judging will be based on percentage of solution completion and elegance in design. 

Interested in participating? 

Frequently Asked Questions


What is The Machine Learning Competition?

The competition will focus on a solution for anaphora in linguistics.  Anaphora represents a challenge for natural language processing (NLP) and named entity recognition (NER) concepts in computational linguistics, and since Kingland’s Cognitive/AI solutions are centered around successful usage of NLP and NER within the field of Text Analytics, having a solution for anaphora is important.

Entrants to the competition are encouraged to use open source NLP and NER technology paired with open source and/or academia-based corpus sets to help with language processing and model training. It is expected that good design and custom coding beyond the use of this open source technology will be required to provide a solution. The English language will be the focus for the competition.

Who can participate?
Open to any Iowa State University undergrad and grad student.

How is the contest conducted?
Online – you will get all information sent to you and you can submit your answer via email. File should be 10 MB or less.

Can I participate with a team?
Yes, students are encouraged to team up to solve the anaphora challenge. Teams of 2-4 students are likely optimal, but you can also participate as an individual.

Can I submit a project I’ve worked on in the past?
Out of the fairness to other participants, all submissions must have no work dedicated to them before the contest begins.

Will I have sign a waiver?
Yes, a waiver will be sent to you to sign before you submit your solution.

I have a friend that wants to participate, how do they sign up?
You can send them to this page to fill out the form. The deadline to register is December 15.

Important dates to keep in mind
Deadline to sign up:  December 15, 2018
Receive additional problem details:  December 21, 2018
Submit solutions:  TBD, likely in February 2019
Winner announced:  TBD, likely in March 2019

What is Anaphora in Linguistics?
In linguisticsanaphora is the use of an expression whose interpretation depends upon another expression in context (its antecedent or postcedent). In a narrower sense, anaphora is the use of an expression that depends specifically upon an antecedent expression and thus is contrasted with cataphora, which is the use of an expression that depends upon a postcedent expression. The anaphoric (referring) term is called an anaphor.

In particular, for this competition, we want to associate pronouns with their antecedents (or postcedents, as the case may be). In programming terms, pronouns are like pointers that point to concepts that are defined explicitly by names elsewhere in the text. For example, in the sentence “Mary gave Shele a contact, and she followed up on it,” the word “she” is a pronoun (an anaphor) and “Shele” is its antecedent. This mapping from “she” to “Shele” is obvious enough to someone reading or hearing the sentence, but to a linguistic model, this association is not at all obvious.

For our problem, we want a model that can recover these mappings from pronoun to antecedent/postcedent with the highest possible F1 score. We will provide a corpus for training and evaluating models. One possible corpus is this, but any corpus that is free for research use should suffice for this purpose.

How much does it cost to participate?
The Machine Learning Competition is free.  

Whom can I expect the communication about the event to come from?
All communication will come from our Recruitment & Talent Coordinator, Shele Blum. Please reach out to her with any questions. Shele.blum@kingland.com 

I can’t participate anymore, how do I let you know?
Please reach out to Shele Blum if you can no longer participate.

When are the winners announced and what do they receive?
As a result, the team or individuals with the best solution will win. There will be over $2,500 in cash prizes. Kingland’s Cognitive/AI engineers will judge the solutions based on elegance of design and their effectiveness in resolving the anaphora challenge on a number of different text/documentation examples. In an absence of a complete or working solution, judging will be based on percentage of solution completion and elegance in design.


Keep checking back for updated information.