Giters

skypea / openie-standalone

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Open IE Build Status

This project contains the principal Open Information Extraction (Open IE) system from the University of Washington (UW). An Open IE system runs over sentences and creates extractions that represent relations in text. For example, consider the following sentence.

The U.S. president Barack Obama gave his speech on Tuesday to thousands of people.

There are many binary relations in this sentence that can be expressed as a triple (A, B, C) where A and B are arguments, and C is the relation between those arguments. Since Open IE is not aligned with an ontology, the relation is a phrase of text. Here is a possible list of the binary relations in the above sentence:

(Barack Obama, is the president of, the U.S.)
(Barack Obama, gave, his speech)
(Barack Obama, gave his speech, on Tuesday)
(Barack Obama, gave his speech, to thousands of people)

The first extraction in the above list is a "noun-mediated extraction", because the extraction has a relation phrase is described by the noun "president". The other extractions are very similar. In fact, they can be represented more informatively as an n-ary extraction. An n-ary extraction can have 0 or more secondary arguments. Here is a possible list of the n-ary relations in the sentence:

(Barack Obama, is the president of, the U.S.)
(Barack Obama, gave, [his speech, on Tuesday, to thousands of people])

Extractions can include more than just the arguments and relation as well. For example, we might be interested in whether the extraction is a negative assertion or a positive assertion, or if it is conditional in some way. Consider the following sentence:

Some people say Barack Obama was born in Kenya.

We would not want to extract that (Barack Obama, was born, in Kenya) alone because this is not true. However, if we have the condition as well, we can have a correct extraction.

Some people say:(Barack Obama, was born in, Kenya)

To see an example of Open IE being used, please visit http://openie.cs.washington.edu/.

Notifications

  • June 2016

This repository is a modification of https://github.com/knowitall/openie at 9dcbf4b0a4fd088d780a3f4480ce9ce811295f30 to run on Scala 2.11.

This project currently has many dependencies that would need to migrate to Scala 2.11. To make the migration and future maintenence easier, @schmmd copied in the dependent libraries:

In addition, some code has been removed to ease migration to Scala 2.11:

Finally, ClearSrl.scala was adjusted to produce consistent behavior in Scala 2.11:

Google Group

Research

Open IE 4.x is the successor to Ollie. Whereas Ollie used bootstrapped dependency parse paths to extract relations (see Open Language Learning for Information Extraction), Open IE 4.x uses similar argument and relation expansion heuristics to create Open IE extractions from SRL frames. Open IE 4.x also extends the defintion of Open IE extractions to include n-ary extractions (extractions with 0 or more arguments 2s).

Buiding

openie uses java-7-openjdk & the sbt build system, so downloading dependencies and compiling is simple. Just run:

sbt compile

Memory requirements

openie requires substantial memory. sbt is configured to use these options by default:

-Xmx4G -XX:+UseConcMarkSweepGC

Usage

Command Line Interface

Check out this project and use sbt to run the CLI:

% git clone git@github.com:allenai/openie-standalone.git
% cd openie-standalone
% sbt "runMain edu.knowitall.openie.OpenIECli"
...
[info] * * * * * * * * * * * * *
[info] * OpenIE 4.1.x is ready *
[info] * * * * * * * * * * * * *

openie takes a number of command line arguments. To see them all run sbt "runMain edu.knowitall.openie.OpenIECli --usage". Of particular interest are --ignore-errors which continues running even if an exception is encountered, --binary which gives the binary(triples) output and --split which splits the input document text into sentences.

There are two output formats in the CLI: a simple format made for ease of reading and a columnated format used for machine processing. The format can be specified with either --format simple or --format column. The simple format is chosen by default.

Code interface

Superset of imports needed for the following examples:

import edu.knowitall.openie.OpenIE
import edu.knowitall.openie.OpenIECli.{ColumnFormat, SimpleFormat}
import java.io.{PrintWriter, StringWriter}

openie.extract(...) returns a sequence of Instance objects, which are containers for extractions and confidences. Each extraction has various fields you can access manually.

toString format

In the simplest case, you can ask each Instance object to produce a string representation of itself:

val sentence = "U.S. President Obama gave a speech"
val instances = openie.extract(sentence)
var s = new StringBuilder()
for (instance <- instances) {
  s.append("Instance: " + instance.toString() + "\n")
}
println(s.toString())

That wIll produce this output:

Instance: 0.93 (U.S. President Obama; gave; a speech)
Instance: 0.88 (Obama; [is] President [of]; United States)

SimpleFormat

OpenIECli provides a SimpleFormat helper that produces similar output:

val sentence = "U.S. President Obama gave a speech"
val instances = openie.extract(sentence)
val sw = new StringWriter()
SimpleFormat.print(new PrintWriter(sw), sentence, instances)
println(sw.toString())

That will produce this output:

U.S. President Obama gave a speech
0.93 (U.S. President Obama; gave; a speech)
0.88 (Obama; [is] President [of]; United States)

ColumnFormat

OpenIECli also provides a ColumnFormat helper that produces output with tabs:

val sentence = "U.S. President Obama gave a speech"
val instances = openie.extract(sentence)
val sw = new StringWriter()
ColumnFormat.print(new PrintWriter(sw), sentence, instances)
println(sw.toString())

That will produce this output:

0.9329286852051247              SimpleArgument(U.S. President Obama,List([0, 20)))      Relation(gave,List([21, 25)))   SimpleArgument(a speech,List([26, 34))) U.S. President Obama gave a speech
0.8847999636040884              SimpleArgument(Obama,List([15, 20)))    Relation([is] President [of],List([5, 14)))     SimpleArgument(United States,List([0, 4)))      U.S. President Obama gave a speech

Usage at AllenAI

Built jars of openie-standalone are published to AllenAI's private repository in Bintray. To use it, add the private repository to the list of dependency resolvers.

import org.allenai.plugins.CoreDependencies

object Dependencies extends CoreDependencies {
    ...
    resolvers += "AllenAI BintrayPrivate" at "http://dl.bintray.com/allenai/private"
    ...
}

Then take a depenency on the latest version of OpenIE:

libraryDependencies ++= Seq(
  ...
  "org.allenai.openie" %% "openie" % "4.2.4"
)

Contributors

About

License:Other

Languages

Language:Scala 99.6%Language:Shell 0.4%