CAT - Content Assignment Tool

What is CAT?

Content Assignment Tool (CAT) was developed at the Center For Natural Language Processing (CNLP) for the CASAA project. The CASAA Project is funded by NSF's NSDL program. CAT automatically suggests appropriate standards for learning resources such as lesson plans, and has functionalities useful for collection builders, catalogers, and curriculum developers.

Content Assignment Tool (CAT)

CAT is a Tool that assists collection providers, catalogers and teachers in assigning content standards by providing suggestions of relevant standards. The Tool makes suggestions which are reviewed by the individual adding the resource to the NSDL. The user then selects and approves the final assignment. As well, if the individual adding the resource would like to consider other standards, the system allows the user to browse a hierarchical list of available standards and select and add their preferred standards to the list of suggested standards. The system stores these human-vetted assignments and utilizes machine learning techniques in order to improve its own auto-suggestions. This learning can take place at the single cataloger level, or at the organizational level where all assignments from multiple catalogers in an organization are aggregated.

CAT Features

NLP

CAT and SAT use Natural Language Processing (NLP) on the content of the educational resources and standards. Through NLP, the system is able to process both the standard and the educational resource with part-of-speech tags, and then bracket verb and noun phrases, categorize terms and phrases, and find synonyms for them in order to facilitate more accurate matching between the resource and the standards.

User Relevance Feedback "More Like This"

Once the system has provided suggestions, the user can select one or more standards from the results and, if they wish, ask the system to find more standards similar to those selected.

Facilitated Standards Navigation "Browse Standards"

Typically, there are a vast number of standards in a single standards document such as the National Council of Teachers of Mathematics and National Science Education Standards. Navigation within this document is cumbersome at best. Our Tools provide an easily navigable standards tree for users to browse the standards and ease their cognitive effort.

Machine Learning

We further utilize the human-vetted assignments in auto-teaching CAT and SAT how to make better assignments. Thus CAT incorporates a continuous assignment quality improvement through a custom built instance-based machine learning algorithm.

Keyword Boosting

A user can also provide additional keywords beyond those found in the content of the educational resource to enhance the retrieval process.

CNLP developed a CAT API which the DLESE Program Center has integrated with the DLESE Collection System (DCS). A demo of this integration can be found here.

We utilize JES & Co.’s ASN XML representation of all existing Math and Science state and national standards in our tool, and use Natural Language Processing technology to provide automatic, intelligent assignments.

CAT and Contact Information

If you are interested in trying out CAT or learning more about it, please contact Anne Diekema - diekemar AT syr.edu or go directly to the CAT tool.

SAT - Standards Alignment Tool

What is SAT?

Standards Alignment Tool (SAT)

SAT creates alignments between state and national standards. Focus groups with teachers have shown that the new generation of teachers prefers to search by state standard for educational resources to aid in their teaching. Teachers stated that state standards were more relevant to their work than national standards because they have to teach to state standards and document student progress in relation to those standards. However, to avoid assigning state standards for 50 different states, most lesson plan repositories, if they catalog using content standards at all, prefer to use the more general national content standards so as not to limit their work to a single state. Manually adding the detailed standards is significantly more time consuming than cataloging at the general level, hence our motivation in developing a computer-assisted methodology.

To create an automatic mapping between state and various national standards, SAT uses a relatively small set of manually-determined alignments between standards to learn classifiers for a crosswalk standard and uses these learned classifiers to align new standards to the crosswalk, thus allowing alignments between any state. This mapping from state standards to national standards can be incorporated into the search capabilities of educational resource repositories so that teachers can search for resources using either their home state standards or national standards. Content standardsÕ assignment will be enhanced by the utilization of the crosswalk because a cataloger can assign standards using their state standards while being assured that the learning resource will still be searchable and ÔfindableÕ using other state standards through use of the crosswalk. As a result, educational resources from anywhere in the country can easily be shared once this translation between state standards is facilitated.

SAT Contact Information

A beta version of SAT with a REST API is currently being tested. If you are interested in trying out SAT or learning more about it, please contact Anne Diekema - diekemar AT syr.edu.