[Web Seminar] Tips for Successful Data Annotation Management to Lead AI Machine Learning ~ From Experience Creating Over 48 Million Data Annually ~ (10/8)

September 11, 2020

[Web Seminar] Tips for Successful Data Annotation Management to Lead AI Machine Learning ~ From Experience Creating Over 48 Million Data Annually ~ (10/8)

AI Machine Learning Data Annotation Seminar on Thursday, October 8th, 2020

 

[Web Seminar] Tips for Successful Data Annotation Management for AI Machine Learning

From over 48 million data creation experiences per year~


Will be held.


———————————————————————————————————————


In machine learning for AI, the quality of annotated data greatly affects the accuracy of AI.

In order to improve accuracy, it is necessary to have a large number of human resources for creating a large amount of training data, making quality and personnel management extremely important.

 

    Data annotation is important not only for the tools and methods used, but also because there are limited areas that can be automated and manual work is unavoidable.

    Therefore, the weight of the data annotator's characteristics and abilities is significant, and no matter how much manualization or standardization is done, if the management point is off, productivity and quality will not improve.

    Therefore, the quality and productivity of data annotation are greatly influenced by the balance between improving human performance and "appropriate balance" of manuals, which is often overlooked.


    At Human Science Co., Ltd., we handle the creation of over 48 million teacher data annually.

    In this seminar, we will focus on the quality management aspect of data annotation based on our expertise, and highlight important points.


  • ① How to incorporate information into manuals
  • Mechanisms and systems that allow for maximum performance from two individuals

    • We will introduce examples and other information.


    This is a seminar that we would like to invite those in charge who have the following challenges to participate in!


  • - Created manuals and guidelines, but not sure how much information to incorporate (there is no end to it once you start incorporating)
  • ・Even though I created the manual, the number of questions and exchanges from data annotators has not decreased.
  • ・Data annotator is struggling to balance quality and productivity.
  • ・Unable to manage a large number of data annotators
  • ・Data annotators are causing large variations in productivity, and I am struggling to improve it even with guidance.
  • ・I cannot manage data annotators (data annotation) because I want to focus on development tasks.

    • After the seminar, we will be happy to provide individual consultations as needed.

    Please check the "Individual Consultation Request" box on the application form.

     

    Feel free to join us!

Program

  • - Seminar Main Session (50 minutes)
  • - Q&A (10 minutes)
  • - Individual Consultation Session (Limited to first 2 applicants)

Target Audience

  • ・Those who are working on AI implementation, PoC, and other data annotation projects in their own company
  • ・Looking into BPO for data annotation
  • ・Those who are experiencing challenges with our data annotation process
  • ・Those who are managing data annotation

 

  • ※If there are more applications than the capacity, we may conduct a lottery.
  • ※We only accept up to 2 participants from the same company.
  • ※If there are a large number of participants, we will hold a lottery.

Event Details

Date and Time
  • October 8, 2020 (Thursday) 16:00-17:00
  • ※Individual Consultation (for those who wish): 5:00 PM~ (Conducted via web conference)
Cost
Free
How to Participate
  • This seminar is a Web seminar using zoom.
    The participation method will be notified by email to those who have applied.
  • For information on the playback environment for zoom, please check here.

Apply Now

  • Please apply for participation in this seminar from the following.
  • A response will be sent from the person in charge.

 

Application Deadline: October 5, 2020 (Monday)

 

  • ※Participation by other companies and individuals may be declined.
  • ※The event may be cancelled if the minimum number of participants is not reached.
  • ※We will contact you about 1 week before the cancellation from the person in charge.
  • ※The method for viewing the web seminar and the access URL will be notified to the participants by email a few days before the event.
  • ※Individual consultations will be conducted in a web conference format.

Instructor

IMG_3577_3

Suzuki Yasuhiro

  • ・Annotation Project Manager
  • ・Engaged in management such as education of domestic and overseas members, performance management, etc. as a leader of the previous job IT global help desk, and involved in document creation, improvement and translation of data center overseas transfer, etc.
  • ・In my current position, I have utilized my experience and knowledge in translation and document work to work as a Localization Project Manager and Consultant, with a focus on machine translation. Currently, I am responsible for managing large-scale natural language data annotation projects (including English to Japanese localization) at client sites.

Kitada Gaku

  • ・Annotation Project Manager
  • ・From the establishment of the Annotation Department, I have gained experience in numerous data annotation tasks and management, from natural language processing to on-site large-scale projects, including client on-site POC projects and remote projects.
  • ・Project management and team leadership, numerous measures have been planned and implemented to activate communication and knowledge sharing among data annotators.
  • ・Currently in charge of managing image data annotation for the automotive industry.

Apply Now

  • Please apply for participation in this seminar from the following.
  • A response will be sent from the person in charge.

 

Application Deadline: October 5, 2020 (Monday)

 

  • ※Participation by other companies and individuals may be declined.
  • ※The event may be cancelled if the minimum number of participants is not reached.
  • ※We will contact you about 1 week before the cancellation from the person in charge.
  • ※The method for viewing the web seminar and the access URL will be notified to the participants by email a few days before the event.
  • ※Individual consultations will be conducted in a web conference format.

Upcoming Seminars


Details will be posted at a later date. If you would like to receive seminar information emails, please register using the button below.



Contact Us

If you have any questions about the content of this seminar, please contact hsweb_inquiry@science.co.jp.

Please contact us at the following number for inquiries by phone.

TEL：03-5321-3111

Contact: Business Development Department Saga