def assignment_c_simplesearchengine_1():
# Use these throughout below.
normalizer = BrainDeadNormalizer()
tokenizer = BrainDeadTokenizer()
# Load and index MeSH terms.
print("LOADING...")
corpus = InMemoryCorpus("../data/mesh.txt")
print("INDEXING...")
inverted_index = InMemoryInvertedIndex(corpus, ["body"], normalizer,
tokenizer)
# Do ranked retrieval, using a simple ranker.
engine = SimpleSearchEngine(corpus, inverted_index)
simple_ranker = BrainDeadRanker()
results = []
# Callback for receiving matches.
def match_collector(match):
results.append(match)
print("*** WINNER", match["score"], match["document"])
query = "polluTION Water"
for match_threshold in [0.1, 1.0]:
print(
f"SEARCHING for '{query}' with match threshold {str(match_threshold)}..."
)
results.clear()
options = {
"match_threshold": match_threshold,
"hit_count": 10,
"debug": False
}
engine.evaluate(query, options, simple_ranker, match_collector)
assert len(results) == {0.1: 10, 1.0: 3}[match_threshold]
for (score, document_id) in [(match["score"],
match["document"].document_id)
for match in results[:3]]:
assert score == 2.0 # Both 'pollution' and 'water'.
assert document_id in [25274, 25275, 25276]
for score in [match["score"] for match in results[3:]]:
assert score == 1.0 # Only 'pollution' or 'water', but not both.
def assignment_d_betterranker():
# Use these throughout below.
normalizer = BrainDeadNormalizer()
tokenizer = BrainDeadTokenizer()
results = []
hit_count = 10
# Callback for receiving matches.
def match_collector(match):
results.append(match)
print("*** WINNER", match["score"], match["document"])
# Load and index some English news sentences. Look at the output and compare the two rankers!
# The naive ranker assigns equal weight to all words (including stopwords), whereas the improved
# ranker does not. The test below for the improved ranker (with document #24 being the winner)
# assumes a straightforward implementation of a TF-IDF ranking scheme as described in the
# textbook.
print("LOADING...")
corpus = InMemoryCorpus("data/en.txt")
print("INDEXING...")
inverted_index = InMemoryInvertedIndex(corpus, ["body"], normalizer,
tokenizer)
simple_ranker = BrainDeadRanker()
better_ranker = BetterRanker(corpus, inverted_index)
engine = SimpleSearchEngine(corpus, inverted_index)
for query in ["the terrorism attack and obama"]:
options = {
"match_threshold": 0.1,
"hit_count": hit_count,
"debug": False
}
for ranker in [simple_ranker, better_ranker]:
print("SEARCHING for '" + query + "' using " +
ranker.__class__.__name__ + "...")
results.clear()
engine.evaluate(query, options, ranker, match_collector)
winner_document_ids = {
simple_ranker: [9221, 7263],
better_ranker: [24]
}[ranker]
assert 0 < len(results) <= hit_count
assert results[0]["document"].document_id in winner_document_ids
def assignment_d_shinglegenerator_2():
# Use these throughout below.
normalizer = BrainDeadNormalizer()
tokenizer = ShingleGenerator(3)
ranker = BrainDeadRanker()
results = []
hit_count = 10
# Load MeSH terms.
print("LOADING...")
corpus = InMemoryCorpus("data/mesh.txt")
# Do ranked retrieval, using n-grams (shingles) and a simple ranker. This allows for fuzzy retrieval.
print("INDEXING...")
inverted_index = InMemoryInvertedIndex(corpus, ["body"], normalizer,
tokenizer)
engine = SimpleSearchEngine(corpus, inverted_index)
# Callback for receiving matches.
def match_collector(match):
results.append(match)
print("*** WINNER", match["score"], match["document"])
# Test with some mispelled queries. Be robust for arbitrary resolving of ties.
for (query, winner_score, winner_document_ids) in [
("orGAnik kEMmistry", 8.0, [16981, 16980, 4411, 4410, 4408]),
("synndrome", 7.0, [1275])
]:
print("SEARCHING for '" + query + "'...")
results.clear()
options = {
"match_threshold": 0.1,
"hit_count": hit_count,
"debug": False
}
engine.evaluate(query, options, ranker, match_collector)
assert 0 < len(results) <= hit_count
assert results[0]["score"] == winner_score
assert results[0]["document"].document_id in winner_document_ids
def assignment_c_simplesearchengine_3():
# All accesses to posting lists are logged here.
accesses = []
# For testing.
class AccessLoggedIterator(Iterator[Posting]):
def __init__(self, term: str, wrapped: Iterator[Posting]):
self._term = term
self._wrapped = wrapped
def __next__(self):
posting = next(self._wrapped)
accesses.append((self._term, posting.document_id))
return posting
# For testing.
class AccessLoggedInvertedIndex(InvertedIndex):
def __init__(self, wrapped: InvertedIndex):
self._wrapped = wrapped
def get_terms(self, buffer: str) -> Iterator[str]:
return self._wrapped.get_terms(buffer)
def get_postings_iterator(self, term: str) -> Iterator[Posting]:
return AccessLoggedIterator(
term, self._wrapped.get_postings_iterator(term))
def get_document_frequency(self, term: str) -> int:
return self._wrapped.get_document_frequency(term)
# Use these throughout below.
normalizer = BrainDeadNormalizer()
tokenizer = BrainDeadTokenizer()
# Load and index MeSH terms.
corpus = InMemoryCorpus("data/mesh.txt")
inverted_index = AccessLoggedInvertedIndex(
InMemoryInvertedIndex(corpus, ["body"], normalizer, tokenizer))
# Do ranked retrieval, using a simple ranker.
engine = SimpleSearchEngine(corpus, inverted_index)
simple_ranker = BrainDeadRanker()
query = "Water polluTION"
options = {"match_threshold": 0.5, "hit_count": 1, "debug": False}
engine.evaluate(query, options, simple_ranker, lambda m: m)
# Expected posting list traversal ordering if the implementation chooses to evaluate this as "water pollution".
ordering1 = [('water', 3078), ('pollution', 788), ('pollution', 789),
('pollution', 790), ('pollution', 8079), ('water', 8635),
('pollution', 23837), ('water', 9379), ('water', 23234),
('water', 25265), ('pollution', 25274), ('water', 25266),
('water', 25267), ('water', 25268), ('water', 25269),
('water', 25270), ('water', 25271), ('water', 25272),
('water', 25273), ('water', 25274), ('water', 25275),
('pollution', 25275), ('water', 25276), ('pollution', 25276),
('water', 25277), ('water', 25278), ('water', 25279),
('water', 25280), ('water', 25281)]
# Expected posting list traversal ordering if the implementation chooses to evaluate this as "pollution water".
ordering2 = [('pollution', 788), ('water', 3078), ('pollution', 789),
('pollution', 790), ('pollution', 8079), ('water', 8635),
('pollution', 23837), ('water', 9379), ('water', 23234),
('water', 25265), ('pollution', 25274), ('water', 25266),
('water', 25267), ('water', 25268), ('water', 25269),
('water', 25270), ('water', 25271), ('water', 25272),
('water', 25273), ('water', 25274), ('pollution', 25275),
('water', 25275), ('pollution', 25276), ('water', 25276),
('water', 25277), ('water', 25278), ('water', 25279),
('water', 25280), ('water', 25281)]
# Check that the posting lists have been accessed in a way that's consistent with document-at-a-time traversal.
# Be somewhat robust to implementation details. This is a fairly strict test, and advanced (but valid)
# implementations that for some reason do lookaheads or whatever might fail.
assert accesses == ordering1 or accesses == ordering2
def test_document_at_a_time_traversal_mesh_corpus(self):
from typing import Iterator, List, Tuple
import os.path
from invertedindex import Posting, InvertedIndex, InMemoryInvertedIndex
from corpus import InMemoryCorpus
from searchengine import SimpleSearchEngine
from ranking import BrainDeadRanker
class AccessLoggedIterator(Iterator[Posting]):
def __init__(self, term: str, history: List[Tuple[str, int]], wrapped: Iterator[Posting]):
self._term = term
self._history = history
self._wrapped = wrapped
def __next__(self):
posting = next(self._wrapped)
self._history.append((self._term, posting.document_id))
return posting
class AccessLoggedInvertedIndex(InvertedIndex):
def __init__(self, wrapped: InvertedIndex):
self._wrapped = wrapped
self._history = []
def get_terms(self, buffer: str) -> Iterator[str]:
return self._wrapped.get_terms(buffer)
def get_postings_iterator(self, term: str) -> Iterator[Posting]:
return AccessLoggedIterator(term, self._history, self._wrapped.get_postings_iterator(term))
def get_document_frequency(self, term: str) -> int:
return self._wrapped.get_document_frequency(term)
def get_history(self) -> List[Tuple[str, int]]:
return self._history
corpus = InMemoryCorpus(os.path.join(data_path, 'mesh.txt'))
index = AccessLoggedInvertedIndex(InMemoryInvertedIndex(corpus, ["body"], self._normalizer, self._tokenizer))
engine = SimpleSearchEngine(corpus, index)
ranker = BrainDeadRanker()
query = "Water polluTION"
options = {"match_threshold": 0.5, "hit_count": 1, "debug": False}
engine.evaluate(query, options, ranker, lambda m: None)
history = index.get_history()
ordering1 = [('water', 3078), # Document-at-a-time ordering if evaluated as "water pollution".
('pollution', 788), ('pollution', 789), ('pollution', 790), ('pollution', 8079),
('water', 8635),
('pollution', 23837),
('water', 9379), ('water', 23234), ('water', 25265),
('pollution', 25274),
('water', 25266), ('water', 25267), ('water', 25268), ('water', 25269), ('water', 25270),
('water', 25271), ('water', 25272), ('water', 25273), ('water', 25274), ('water', 25275),
('pollution', 25275),
('water', 25276),
('pollution', 25276),
('water', 25277), ('water', 25278), ('water', 25279), ('water', 25280), ('water', 25281)]
ordering2 = [('pollution', 788), # Document-at-a-time ordering if evaluated as "pollution water".
('water', 3078),
('pollution', 789), ('pollution', 790), ('pollution', 8079),
('water', 8635),
('pollution', 23837),
('water', 9379), ('water', 23234), ('water', 25265),
('pollution', 25274),
('water', 25266), ('water', 25267), ('water', 25268), ('water', 25269), ('water', 25270),
('water', 25271), ('water', 25272), ('water', 25273), ('water', 25274),
('pollution', 25275),
('water', 25275),
('pollution', 25276),
('water', 25276), ('water', 25277), ('water', 25278), ('water', 25279), ('water', 25280),
('water', 25281)]
self.assertTrue(history == ordering1 or history == ordering2) # Strict. Advanced implementations might fail.
def assignment_d():
# Use these throughout below.
normalizer = BrainDeadNormalizer()
simple_tokenizer = BrainDeadTokenizer()
# Load MeSH terms.
print("LOADING...")
corpus = InMemoryCorpus("data/mesh.txt")
# Do ranked retrieval, using n-grams and a simple ranker. This allows for fuzzy retrieval.
print("INDEXING...")
shingle_generator = ShingleGenerator(3)
shingle_inverted_index = InMemoryInvertedIndex(corpus, ["body"],
normalizer,
shingle_generator)
shingle_engine = SimpleSearchEngine(corpus, shingle_inverted_index)
simple_ranker = BrainDeadRanker()
results = []
hit_count = 10
# Callback for receiving matches.
def match_collector(match):
results.append(match)
print("*** WINNER", match["score"], match["document"])
# Test with some mispelled queries. Be robust for arbitrary resolving of ties.
for (query, winner_score, winner_document_ids) in [
("orGAnik kEMmistry", 8.0, [16981, 16980, 4411, 4410, 4408]),
("synndrome", 7.0, [1275])
]:
print("SEARCHING for '" + query + "'...")
results.clear()
options = {
"match_threshold": 0.1,
"hit_count": hit_count,
"debug": False
}
shingle_engine.evaluate(query, options, simple_ranker, match_collector)
assert 0 < len(results) <= hit_count
assert results[0]["score"] == winner_score
assert results[0]["document"].document_id in winner_document_ids
# Load and index some English news sentences. Look at the output and compare the two rankers!
# The naive ranker assigns equal weight to all words (including stopwords), whereas the improved
# ranker does not. The test below for the improved ranker (with document #24 being the winner)
# assumes a straightforward implementation of a TF-IDF ranking scheme as described in the
# textbook.
print("LOADING...")
corpus = InMemoryCorpus("data/en.txt")
print("INDEXING...")
inverted_index = InMemoryInvertedIndex(corpus, ["body"], normalizer,
simple_tokenizer)
better_ranker = BetterRanker(corpus, inverted_index)
engine = SimpleSearchEngine(corpus, inverted_index)
for query in ["the terrorism attack and obama"]:
options = {
"match_threshold": 0.1,
"hit_count": hit_count,
"debug": False
}
for ranker in [simple_ranker, better_ranker]:
print("SEARCHING for '" + query + "' using " +
ranker.__class__.__name__ + "...")
results.clear()
engine.evaluate(query, options, ranker, match_collector)
winner_document_ids = {
simple_ranker: [9221, 7263],
better_ranker: [24]
}[ranker]
assert 0 < len(results) <= hit_count
assert results[0]["document"].document_id in winner_document_ids
def test_simple_search_daat():
from typing import Iterator
from ranking import BrainDeadRanker
from searchengine import SimpleSearchEngine
from invertedindex import InvertedIndex, Posting
from corpus import Corpus, Document, InMemoryDocument
inverted_index = None
class DaatPostingList:
def __init__(self, docid_list):
self.docid_list = docid_list
self.idx = 0
@property
def current_docid(self):
if self.idx >= len(self.docid_list):
return 1000
return self.docid_list[self.idx]
def __iter__(self):
return self
def __next__(self):
if self.idx + 1 >= len(self.docid_list):
raise StopIteration
self.idx += 1
docid = self.docid_list[self.idx]
if 106 <= docid <= 108:
maxdiff = 0
for pl in (e for e in inverted_index.active_lists
if e is not self):
maxdiff = max(maxdiff, abs(pl.current_docid - docid))
if maxdiff > 20:
inverted_index.is_daat = False
return Posting(docid, 1)
class DaatCorpus(Corpus):
def __iter__(self):
return (self.get_document(i) for i in range(0, self.size()))
def size(self) -> int:
return 300
def get_document(self, document_id: int) -> Document:
return InMemoryDocument(
document_id,
{'body': inverted_index.get_doc_body(document_id)})
class DaatInvertedIndex(InvertedIndex):
def __init__(self):
self.is_daat = True
self.active_lists = []
_data = {
'a': (1, 2, 7, 10, 11, 12, 20, 21, 22, 41, 42, 44, 45, 50, 100,
103, 106, 109, 112, 201, 203, 204, 205, 206, 207),
'b': (3, 4, 8, 16, 17, 18, 20, 22, 31, 41, 43, 101, 104, 107, 110,
113, 202, 204),
'c': (5, 6, 9, 13, 14, 15, 20, 21, 31, 41, 102, 105, 108, 111, 114,
203, 204, 208),
}
@classmethod
def get_doc_body(cls, document_id):
return ' '.join(k for k, a in cls._data.items()
if document_id in a)
def get_terms(self, buffer: str) -> Iterator[str]:
return iter(buffer.split(' '))
def get_postings_iterator(self, term: str) -> Iterator[Posting]:
if term in self._data:
iterator = DaatPostingList(self._data[term])
self.active_lists.append(iterator)
return iterator
return iter(tuple())
def get_document_frequency(self, term: str) -> int:
return len(self._data.get(term, []))
inverted_index = DaatInvertedIndex()
engine = SimpleSearchEngine(DaatCorpus(), inverted_index)
results = []
def match(m):
results.append((m['score'], m['document'].document_id))
engine.evaluate('a b c', {
'recall_threshold': 1.0,
'hit_count': 100
}, BrainDeadRanker(), match)
results.sort()
if results != [(3.0, 20), (3.0, 41), (3.0, 204)]:
print("FAILURE: search failed.")
if not inverted_index.is_daat:
print("WARNING: not performing proper Document-At-A-Time search!")