← home
Github: datasets/highwire.py

ir_datasets: Highwire (TREC Genomics 2006-07)

Index
  1. highwire
  2. highwire/trec-genomics-2006
  3. highwire/trec-genomics-2007

"highwire"

Medical document collection from Highwire Press. Includes 162,259 scientific articles from 49 journals.

This dataset is used for the TREC 2006-07 TREC Genomics track.

Note that these documents are split into passages based on paragraph tags in the HTML.

docs

Language: en

Document type:
HighwireDoc: (namedtuple)
  1. doc_id: str
  2. journal: str
  3. title: str
  4. spans: Tuple[
    HighwireSpan: (namedtuple)
    1. start: int
    2. length: int
    3. text: str
    , ...]

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("highwire")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, journal, title, spans>

You can find more details about the Python API here.

CLI 
ir_datasets export highwire docs
[doc_id]    [journal]    [title]    [spans]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:highwire')
# Index highwire
indexer = pt.IterDictIndexer('./indices/highwire')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['journal', 'title'])

You can find more details about PyTerrier indexing here.

"highwire/trec-genomics-2006"

The TREC Genomics Track 2006 benchmark. Contains 28 queries with passage-level relevance judgments.

queries

Language: en

Query type:
GenericQuery: (namedtuple)
  1. query_id: str
  2. text: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("highwire/trec-genomics-2006")
for query in dataset.queries_iter():
    query # namedtuple<query_id, text>

You can find more details about the Python API here.

CLI 
ir_datasets export highwire/trec-genomics-2006 queries
[query_id]    [text]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:highwire/trec-genomics-2006')
index_ref = pt.IndexRef.of('./indices/highwire') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pipeline(dataset.get_topics())

You can find more details about PyTerrier retrieval here.

docs

Language: en

Note: Uses docs from highwire

Document type:
HighwireDoc: (namedtuple)
  1. doc_id: str
  2. journal: str
  3. title: str
  4. spans: Tuple[
    HighwireSpan: (namedtuple)
    1. start: int
    2. length: int
    3. text: str
    , ...]

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("highwire/trec-genomics-2006")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, journal, title, spans>

You can find more details about the Python API here.

CLI 
ir_datasets export highwire/trec-genomics-2006 docs
[doc_id]    [journal]    [title]    [spans]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:highwire/trec-genomics-2006')
# Index highwire
indexer = pt.IterDictIndexer('./indices/highwire')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['journal', 'title'])

You can find more details about PyTerrier indexing here.

qrels
Query relevance judgment type:
HighwireQrel: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. start: int
  4. length: int
  5. relevance: int

Relevance levels

Rel.Definition
0NOT
1POSSIBLY
2DEFINITELY

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("highwire/trec-genomics-2006")
for qrel in dataset.qrels_iter():
    qrel # namedtuple<query_id, doc_id, start, length, relevance>

You can find more details about the Python API here.

CLI 
ir_datasets export highwire/trec-genomics-2006 qrels --format tsv
[query_id]    [doc_id]    [start]    [length]    [relevance]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
from pyterrier.measures import *
pt.init()
dataset = pt.get_dataset('irds:highwire/trec-genomics-2006')
index_ref = pt.IndexRef.of('./indices/highwire') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pt.Experiment(
    [pipeline],
    dataset.get_topics(),
    dataset.get_qrels(),
    [MAP, nDCG@20]
)

You can find more details about PyTerrier experiments here.

Citation
bibtex: @inproceedings{Hersh2006TrecGenomics, title={TREC 2006 Genomics Track Overview}, author={William Hersh and Aaron M. Cohen and Phoebe Roberts and Hari Krishna Rekapalli}, booktitle={TREC}, year={2006} }

"highwire/trec-genomics-2007"

The TREC Genomics Track 2007 benchmark. Contains 36 queries with passage-level relevance judgments.

queries

Language: en

Query type:
GenericQuery: (namedtuple)
  1. query_id: str
  2. text: str

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("highwire/trec-genomics-2007")
for query in dataset.queries_iter():
    query # namedtuple<query_id, text>

You can find more details about the Python API here.

CLI 
ir_datasets export highwire/trec-genomics-2007 queries
[query_id]    [text]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:highwire/trec-genomics-2007')
index_ref = pt.IndexRef.of('./indices/highwire') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pipeline(dataset.get_topics())

You can find more details about PyTerrier retrieval here.

docs

Language: en

Note: Uses docs from highwire

Document type:
HighwireDoc: (namedtuple)
  1. doc_id: str
  2. journal: str
  3. title: str
  4. spans: Tuple[
    HighwireSpan: (namedtuple)
    1. start: int
    2. length: int
    3. text: str
    , ...]

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("highwire/trec-genomics-2007")
for doc in dataset.docs_iter():
    doc # namedtuple<doc_id, journal, title, spans>

You can find more details about the Python API here.

CLI 
ir_datasets export highwire/trec-genomics-2007 docs
[doc_id]    [journal]    [title]    [spans]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
pt.init()
dataset = pt.get_dataset('irds:highwire/trec-genomics-2007')
# Index highwire
indexer = pt.IterDictIndexer('./indices/highwire')
index_ref = indexer.index(dataset.get_corpus_iter(), fields=['journal', 'title'])

You can find more details about PyTerrier indexing here.

qrels
Query relevance judgment type:
HighwireQrel: (namedtuple)
  1. query_id: str
  2. doc_id: str
  3. start: int
  4. length: int
  5. relevance: int

Relevance levels

Rel.Definition
0NOT_RELEVANT
1RELEVANT

Examples:

Python API 
import ir_datasets
dataset = ir_datasets.load("highwire/trec-genomics-2007")
for qrel in dataset.qrels_iter():
    qrel # namedtuple<query_id, doc_id, start, length, relevance>

You can find more details about the Python API here.

CLI 
ir_datasets export highwire/trec-genomics-2007 qrels --format tsv
[query_id]    [doc_id]    [start]    [length]    [relevance]
...

You can find more details about the CLI here.

PyTerrier 
import pyterrier as pt
from pyterrier.measures import *
pt.init()
dataset = pt.get_dataset('irds:highwire/trec-genomics-2007')
index_ref = pt.IndexRef.of('./indices/highwire') # assumes you have already built an index
pipeline = pt.BatchRetrieve(index_ref, wmodel='BM25')
# (optionally other pipeline components)
pt.Experiment(
    [pipeline],
    dataset.get_topics(),
    dataset.get_qrels(),
    [MAP, nDCG@20]
)

You can find more details about PyTerrier experiments here.

Citation
bibtex: @inproceedings{Hersh2007TrecGenomics, title={TREC 2007 Genomics Track Overview}, author={William Hersh and Aaron Cohen and Lynn Ruslen and Phoebe Roberts}, booktitle={TREC}, year={2007} }