Please click on the following button to go to the UVI Tutorial Video:


VerbNet API

The VerbNet API contains python tools for working with VerbNet data (XML files). It also contains the latest publicly available version of VerbNet. The project can be found at


The Unified Verb Index is a system which merges links and web pages from four different natural language processing projects: VerbNet, PropBank, FrameNet, WordNet and OntoNotes. SemLink is a project whose aim is to link together these lexical resources via sets of mappings. These mappings have made it possible to combine the information provided by these different lexical resources for tasks such as inference. The project can be found at

Force Dynamics

Force dynamics is a generalization of the theory of causal relations based on transmission of force; Talmy's extended theory of causation (Talmy 1988/2000; Croft 2012). In the force dynamic model, event structure is defined in terms of the interactions between the participants in the event. An argument structure construction is the configuration of subject, object and oblique phrases which a verb occurs with. (Croft et al., 2018)

Verb classes in VerbNet are linked to Force Dynamics relations where available.

VerbNet Parser

The VerbNet Parser provides a tool that uses VerbNet class predictions and PropBank semantic roles to align to a VerbNet frame and produce VerbNet semantic representations. The demo of the VerbNet Parser can be found here:

This is the link of the VerbNet Parser Github:

Story Realization

Neural network based approaches to automated story plot generation attempt to learn how to generate novel plots from a corpus of natural language plot summaries. The project of Story Realization presents an ensemble-based model that generates natural language guided by events. They provide results---including a human subjects study---for a full end-to-end automated story generation system showing that their method generates more coherent and plausible stories than baseline approaches.VerbNet is used in the experiments.

The paper of this project can be found here: Story Realization: Expanding Plot Events into Sentences


Syn-QG(Syntactic and Shallow Semantic Rules for Question Generation) is a Java-based question generator which generates questions from multiple sources: 1. Dependency Parsing. 2. Semantic Role Labeling. 3. NER templates. 4. VerbNet Predicates. 5. PropBank Roleset Argument Descriptions. 6. Custom Rules. 7. Implication Rules. The paper and code of this project can be found here: Kaustubh Dholé’s Blog


UVI stores lexical resources as well as mappings between the resources. For access to lexical resources used in this project; select the projects/ resources you are interested in and click the Download button. Data is downloaded in JSON Format.

Include Resources:

Other VerbNet Resources:
VerbNet v3.3
VerbNet v3.2
VerbNet v3.1
VerbNet Guidelines