← home
Github: datasets/trec_arabic.py

ir_datasets: TREC Arabic

Index
  1. trec-arabic
  2. trec-arabic/ar2001
  3. trec-arabic/ar2002

Data Access Information

To use this dataset, you need a copy of the source corpus, provided by the the Linguistic Data Consortium. The specific resource needed is LDC2001T55.

Many organizations already have a subscription to the LDC, so access to the collection can be as easy as confirming the data usage agreement and downloading the corpus. Check with your library for access details.

The source file is: arabic_newswire_a_LDC2001T55.tgz.

ir_datasets expects this file to be copied/linked as ~/.ir_datasets/trec-arabic/corpus.tgz.

"trec-arabic"

A collection of news articles in Arabic, used for multi-lingual evaluation in TREC 2001 and TREC 2002.

Document collection from LDC2001T55.

docs
384K docs

Language: ar

Document type:
TrecDoc: (namedtuple)
  1. doc_id: str
  2. text: str
  3. marked_up_doc: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("trec-arabic")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, text, marked_up_doc>

You can find more details about the Python API here.

CLI 
ir_datasets export trec-arabic docs
[doc_id]    [text]    [marked_up_doc]
...

You can find more details about the CLI here.

PyTerrier

No example available for PyTerrier

Citation

ir_datasets.bib:

\cite{Graff2001Arabic}

Bibtex:

@misc{Graff2001Arabic, title={Arabic Newswire Part 1 LDC2001T55}, author={Graff, David, and Walker, Kevin}, year={2001}, url={https://catalog.ldc.upenn.edu/LDC2001T55}, publisher={Linguistic Data Consortium} }
Metadata 
{
  "docs": {
    "count": 383872,
    "fields": {
      "doc_id": {
        "max_len": 21,
        "common_prefix": ""
      }
    }
  }
}

"trec-arabic/ar2001"

Arabic benchmark from TREC 2001.

queries
25 queries

Language: ar

Query type:
TrecQuery: (namedtuple)
  1. query_id: str
  2. title: str
  3. description: str
  4. narrative: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("trec-arabic/ar2001")
for query in dataset.queries_iter():
    query # namedtuple<query_id, title, description, narrative>

You can find more details about the Python API here.

CLI 
ir_datasets export trec-arabic/ar2001 queries
[query_id]    [title]    [description]    [narrative]
...

You can find more details about the CLI here.

PyTerrier

No example available for PyTerrier

docs
384K docs

Inherits docs from trec-arabic

Language: ar

Document type:
TrecDoc: (namedtuple)
  1. doc_id: str
  2. text: str
  3. marked_up_doc: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("trec-arabic/ar2001")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, text, marked_up_doc>

You can find more details about the Python API here.

CLI 
ir_datasets export trec-arabic/ar2001 docs
[doc_id]    [text]    [marked_up_doc]
...

You can find more details about the CLI here.

PyTerrier

No example available for PyTerrier

qrels
23K qrels
Query relevance judgment type:
TrecQrel: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. relevance: int
  4. iteration: str

Relevance levels

Rel.DefinitionCount%
0not relevant19K81.9%
1relevant4.1K18.1%

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("trec-arabic/ar2001")
for qrel in dataset.qrels_iter():
    qrel # namedtuple<query_id, doc_id, relevance, iteration>

You can find more details about the Python API here.

CLI 
ir_datasets export trec-arabic/ar2001 qrels --format tsv
[query_id]    [doc_id]    [relevance]    [iteration]
...

You can find more details about the CLI here.

PyTerrier

No example available for PyTerrier

Citation

ir_datasets.bib:

\cite{Gey2001Arabic,Graff2001Arabic}

Bibtex:

@inproceedings{Gey2001Arabic, title={The TREC-2001 Cross-Language Information Retrieval Track: Searching Arabic using English, French or Arabic Queries}, author={Fredric Gey and Douglas Oard}, booktitle={TREC}, year={2001} } @misc{Graff2001Arabic, title={Arabic Newswire Part 1 LDC2001T55}, author={Graff, David, and Walker, Kevin}, year={2001}, url={https://catalog.ldc.upenn.edu/LDC2001T55}, publisher={Linguistic Data Consortium} }
Metadata 
{
  "docs": {
    "count": 383872,
    "fields": {
      "doc_id": {
        "max_len": 21,
        "common_prefix": ""
      }
    }
  },
  "queries": {
    "count": 25
  },
  "qrels": {
    "count": 22744,
    "fields": {
      "relevance": {
        "counts_by_value": {
          "0": 18622,
          "1": 4122
        }
      }
    }
  }
}

"trec-arabic/ar2002"

Arabic benchmark from TREC 2002.

queries
50 queries

Language: ar

Query type:
TrecQuery: (namedtuple)
  1. query_id: str
  2. title: str
  3. description: str
  4. narrative: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("trec-arabic/ar2002")
for query in dataset.queries_iter():
    query # namedtuple<query_id, title, description, narrative>

You can find more details about the Python API here.

CLI 
ir_datasets export trec-arabic/ar2002 queries
[query_id]    [title]    [description]    [narrative]
...

You can find more details about the CLI here.

PyTerrier

No example available for PyTerrier

docs
384K docs

Inherits docs from trec-arabic

Language: ar

Document type:
TrecDoc: (namedtuple)
  1. doc_id: str
  2. text: str
  3. marked_up_doc: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("trec-arabic/ar2002")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, text, marked_up_doc>

You can find more details about the Python API here.

CLI 
ir_datasets export trec-arabic/ar2002 docs
[doc_id]    [text]    [marked_up_doc]
...

You can find more details about the CLI here.

PyTerrier

No example available for PyTerrier

qrels
38K qrels
Query relevance judgment type:
TrecQrel: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. relevance: int
  4. iteration: str

Relevance levels

Rel.DefinitionCount%
0not relevant33K84.6%
1relevant5.9K15.4%

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("trec-arabic/ar2002")
for qrel in dataset.qrels_iter():
    qrel # namedtuple<query_id, doc_id, relevance, iteration>

You can find more details about the Python API here.

CLI 
ir_datasets export trec-arabic/ar2002 qrels --format tsv
[query_id]    [doc_id]    [relevance]    [iteration]
...

You can find more details about the CLI here.

PyTerrier

No example available for PyTerrier

Citation

ir_datasets.bib:

\cite{Gey2002Arabic,Graff2001Arabic}

Bibtex:

@inproceedings{Gey2002Arabic, title={The TREC-2002 Arabic/English CLIR Track}, author={Fredric Gey and Douglas Oard}, booktitle={TREC}, year={2002} } @misc{Graff2001Arabic, title={Arabic Newswire Part 1 LDC2001T55}, author={Graff, David, and Walker, Kevin}, year={2001}, url={https://catalog.ldc.upenn.edu/LDC2001T55}, publisher={Linguistic Data Consortium} }
Metadata 
{
  "docs": {
    "count": 383872,
    "fields": {
      "doc_id": {
        "max_len": 21,
        "common_prefix": ""
      }
    }
  },
  "queries": {
    "count": 50
  },
  "qrels": {
    "count": 38432,
    "fields": {
      "relevance": {
        "counts_by_value": {
          "0": 32523,
          "1": 5909
        }
      }
    }
  }
}