def effectiveness_tfidf(relevance_file):
# TFIDF baseline run
baseline_run = path.join(path.pardir, "Retrieval", "Retrieval_Model",
"TFIDF_Model", "TFIDF_results.json")
# MAP
v = mean_average_precision(baseline_run, relevance_file)
# MRR
mrr = mean_reciprocal_rank(baseline_run, relevance_file)
# Precision and Recall
precision_recall_calculator(baseline_run, relevance_file, "TF_IDF")
# Precision at rank =5 and rank = 20
precision_file = path.join(path.dirname(path.abspath(__file__)), "Files",
"precision_values_" + "TF_IDF" + ".json")
precision_at_rank_calculator(precision_file, "TF_IDF")
values = result_scores["Mean Average Precision"]
values.append(["TFIDF", v])
values = result_scores["Mean_Reciprocal_Rank"]
values.append(["TFIDF", mrr])
# Stopped TFIDF baseline run
baseline_run = path.join(path.pardir, "Retrieval", "Retrieval_Model",
"TFIDF_Model", "Stopped_TFIDF_results.json")
# MAP
v = mean_average_precision(baseline_run, relevance_file)
# MRR
mrr = mean_reciprocal_rank(baseline_run, relevance_file)
# Precision and Recall
precision_recall_calculator(baseline_run, relevance_file, "TF_IDF_Stopped")
# Precision at rank =5 and rank = 20
precision_file = path.join(path.dirname(path.abspath(__file__)), "Files",
"precision_values_" + "TF_IDF_Stopped" + ".json")
precision_at_rank_calculator(precision_file, "TF_IDF_Stopped")
values = result_scores["Mean Average Precision"]
values.append(["Stopped_TFIDF", v])
values = result_scores["Mean_Reciprocal_Rank"]
values.append(["Stopped_TFIDF", mrr])
print("MAP, MRR and Precision and Recall generated for TFIDF Model.\n")
def effectiveness_bm25(relevance_file):
baseline_run = path.join(path.pardir, "Retrieval", "Retrieval_Model",
"BM25_Model", "bm25_baseline_results.json")
# MAP for BM25 baseline run
v = mean_average_precision(baseline_run, relevance_file)
# MRR for BM25 baseline run
mrr = mean_reciprocal_rank(baseline_run, relevance_file)
# Precision and Recall
precision_recall_calculator(baseline_run, relevance_file, "BM25")
# Precision at rank =5 and rank = 20
precision_file = path.join(path.dirname(path.abspath(__file__)), "Files",
"precision_values_" + "BM25" + ".json")
precision_at_rank_calculator(precision_file, "BM25")
values = result_scores["Mean Average Precision"]
values.append(["BM25", v])
values = result_scores["Mean_Reciprocal_Rank"]
values.append(["BM25", mrr])
# BM25 Stopped baseline run
baseline_run = path.join(path.pardir, "Retrieval", "Retrieval_Model",
"BM25_Model", "bm25_stopped_baseline_results.json")
# MAP
v = mean_average_precision(baseline_run, relevance_file)
# MRR
mrr = mean_reciprocal_rank(baseline_run, relevance_file)
# Precision and Recall
precision_recall_calculator(baseline_run, relevance_file, "BM25_Stopped")
# Precision at rank =5 and rank = 20
precision_file = path.join(path.dirname(path.abspath(__file__)), "Files",
"precision_values_" + "BM25_Stopped" + ".json")
precision_at_rank_calculator(precision_file, "BM25_Stopped")
values = result_scores["Mean Average Precision"]
values.append(["BM25 Stopped", v])
values = result_scores["Mean_Reciprocal_Rank"]
values.append(["BM25 Stopped", mrr])
print("MAP, MRR and Precision and Recall generated for BM25 Model.\n")
def effectiveness_noise_minimized_queries(relevance_file):
baseline_run = path.join(path.pardir, "Extra_Credit", "Noise_Minimizer",
"noise_minimized_bm25_result.json")
# MAP for BM25 on noise-induced queries baseline run
v = mean_average_precision(baseline_run, relevance_file)
# MRR for BM25 on noise-induced queries baseline run
mrr = mean_reciprocal_rank(baseline_run, relevance_file)
# Precision and Recall
precision_recall_calculator(baseline_run, relevance_file, "BM25 on noise-minimized queries")
# Precision at rank =5 and rank = 20
precision_file = path.join(path.dirname(path.abspath(__file__)), "Files",
"precision_values_" + "BM25 on noise-minimized queries" + ".json")
precision_at_rank_calculator(precision_file, "BM25 on noise-minimized queries")
values = result_scores["Mean Average Precision"]
values.append(["BM25 on noise-minimized queries", v])
values = result_scores["Mean_Reciprocal_Rank"]
values.append(["BM25 on noise-minimized queries", mrr])
print("MAP, MRR and Precision and Recall generated for BM25 on noise-minimized queries Model.\n")
def effectiveness_lucene(relevance_file):
# Lucene baseline run
baseline_run = path.join(path.pardir, "Retrieval", "Retrieval_Model",
"LuceneSearchEngine", "result_dict.json")
# MAP
v = mean_average_precision(baseline_run, relevance_file)
# MRR
mrr = mean_reciprocal_rank(baseline_run, relevance_file)
# Precision and Recall
precision_recall_calculator(baseline_run, relevance_file, "Lucene")
# Precision at rank =5 and rank = 20
precision_file = path.join(path.dirname(path.abspath(__file__)), "Files",
"precision_values_" + "Lucene" + ".json")
precision_at_rank_calculator(precision_file, "Lucene")
values = result_scores["Mean Average Precision"]
values.append(["Lucene", v])
values = result_scores["Mean_Reciprocal_Rank"]
values.append(["Lucene", mrr])
print("MAP, MRR and Precision and Recall generated for Lucene Model.\n")
def effectiveness_query_expansion(relevance_file):
baseline_run = path.join(path.pardir, "Retrieval", "Query_Expansion",
"stemmed_baseline_results.json")
# MAP for Query_Expansion baseline run
v = mean_average_precision(baseline_run, relevance_file)
# MRR for Query_Expansion baseline run
mrr = mean_reciprocal_rank(baseline_run, relevance_file)
# Precision and Recall
precision_recall_calculator(baseline_run, relevance_file, "Query_Expansion")
# Precision at rank =5 and rank = 20
precision_file = path.join(path.dirname(path.abspath(__file__)), "Files",
"precision_values_" + "Query_Expansion" + ".json")
precision_at_rank_calculator(precision_file, "Query_Expansion")
result_scores["Mean Average Precision"] = [["Query_expansion", v]]
result_scores["Mean_Reciprocal_Rank"] = [["Query_expansion", mrr]]
print("MAP, MRR and Precision and Recall generated for Query Expansion Model.\n")