Erulemaking cs501 presentation 2
1 / 20

eRulemaking CS501 Presentation 2 - PowerPoint PPT Presentation

  • Uploaded on

eRulemaking CS501 Presentation 2. Who We Are. Sam Phillips MEng in CS Dan Rassi Junior in CS Michael Wang MEng in CS Krzysztof Findeisen Senior in Astro and CS Raymond McGill Senior in IS. Project Overview. Federal Requirement To Read Comments To Proposed Rulemakings

I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
Download Presentation

PowerPoint Slideshow about 'eRulemaking CS501 Presentation 2' - kalea

An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.

- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript
Erulemaking cs501 presentation 2

eRulemakingCS501 Presentation 2

Who we are
Who We Are

  • Sam Phillips

    • MEng in CS

  • Dan Rassi

    • Junior in CS

  • Michael Wang

    • MEng in CS

  • Krzysztof Findeisen

    • Senior in Astro and CS

  • Raymond McGill

    • Senior in IS

Project overview
Project Overview

  • Federal Requirement To Read Comments To Proposed Rulemakings

  • Cornell eRulemaking Initiative (CERI) working on a system to Automatically classify comments.

  • Classification Techniques Need “Supervised Learning”


EARS – Electronic Annotation and Rulemaking System

Will provide a single interface for managing comments the government receives as part of its eRulemaking process

Will use Natural Language Processing (NLP) tools to automate handling of large comment sets

We are working on a prototype EARS for the Legal Information Institute (LII)

Tom Bruce of the LII is our chief contact, but we are also working with several other LII groups

As of Phase II, we had a simple, nonfunctional website that demonstrated our interface

The stakeholders
The Stakeholders

Funding: NSF

Long-Term Users:

Agency Analysts


Other Universities

Grantee: Cornell

eRulemaking Initiative

Subject Matter

Experts: LII Student



NLP Group




Our Group

Term dictionary
Term Dictionary

Rule / Reg.: Proposed rule by a federal agency

Rulemaker / Analyst: Domain expert in agency

Issue: A logical facet which the Rule impacts.

Annotate / Tag (v): To “highlight” text and associate it with a specific issue.

Metadata: Data about Data

(e.g. E-mail to/from/size)

Tag (n): An issue as metadata

Flag (n): Non-issue related metadata (e.g. workflow)

Activities from start to phase ii
Activities From Start To Phase II

  • Meetings With Tom Bruce

    • Introduced Project

    • Explained Requirements / Known Unknowns

  • Meetings With LII Student Annotators Heidi Craig and Laura Klimpel

    • Discussed Current Annotation System

    • Got Feedback for Early Design Ideas

  • Created Static Webpage To Prove That It’s Possible

  • Attended Full CERI Meetings

Activities since phase ii report
Activities Since Phase II Report

  • Creation of Backend / Middleware Architecture

    • Backend in relational mySQL database

    • Middleware in OO PHP

  • Clarification of Some Requirements

    • XML Format

    • Color of highlights

  • Discovery of Some Known Unknowns

    • How NLP System Should React

    • How Extra Data Should Be Displayed

System overview
System Overview

Add / Remove

Rules, Tags,




Choose Rule


Choose Comment

Add / Remove


Design overview
Design Overview

  • Web Site backed by a central database

General design strategy
General Design Strategy

  • Our system architecture is highly modular

    • Website, database, etc. can be swapped out easily

  • All components already available on LII servers

Database design
Database Design

  • Primary goal: flexibility

    • Unified representation of data

    • Supports more than our web release will

    • Lots of room for administrator preferences

  • Secondary goal: speed

    • 4000 regulations issued per year

    • Usually ~100, up to 500,000 comments per regulation

    • Demands on the LII version will be much lower

Web technology
Web Technology

  • Currently using the Drupal Content Management System on LII server to host our web application, however we have minimized this dependence

  • Website uses JavaScript to dynamically change contents of page when user performs an action

  • AJAX technology is used to send annotations between client and server without reloading page

  • Our primary goal has been client compatibility across major browsers and operating systems

Where we re going
Where We’re Going

  • Documentation

    • Describe SQL Scheme and ER Diagram To Future CS501 Groups

      • Include Design Decisions

      • Include mySQL specific queries

    • Describe How Implemented Features Work

      • Low Level (Comments in Code)

      • High Level (Why Features Are Needed / Trade Offs)

    • Describe How Unimplemented Features Might Work

      • Design Considerations

      • Stakeholders Affected

Where we re going 2
Where We’re Going (2)

  • Features

    • Will Certainly Add

      • UI To Add / Remove

        • Comments

        • Rules

        • Metadata Sets

        • Metadata Names

    • Will Fix UI For

      • Deleting Comments

      • Navigating Comments

    • May Add

      • Hierarchical Tags

      • “Fake” NLP Interaction

      • Multi-user Interaction

      • NLP XML To/From Connection

      • Colors