>>> from sklearn.decomposition import PCA
>>> from techminer2.factor_analysis.tfidf import cosine_similarities
>>> cosine_similarities(
... #
... # PARAMS:
... field="author_keywords",
... #
... # TF PARAMS:
... is_binary=True,
... cooc_within=1,
... #
... # TF-IDF PARAMS:
... norm=None,
... use_idf=False,
... smooth_idf=False,
... sublinear_tf=False,
... #
... # TERM PARAMS:
... top_n=20,
... occ_range=(None, None),
... gc_range=(None, None),
... custom_terms=None,
... #
... # DESOMPOSITION PARAMS:
... decomposition_estimator = PCA(
... n_components=5,
... whiten=False,
... svd_solver="auto",
... tol=0.0,
... iterated_power="auto",
... n_oversamples=10,
... power_iteration_normalizer="auto",
... random_state=0,
... ),
... #
... # DATABASE PARAMS:
... root_dir="example/",
... database="main",
... year_filter=(None, None),
... cited_by_filter=(None, None),
... ).head()
cosine_similariries
author_keywords
FINTECH 31:5168 INNOVATION 07:0911 (0.106)
INNOVATION 07:0911 TECHNOLOGY 02:0310 (0.771); BANKING 02:0291 (0...
FINANCIAL_SERVICES 04:0667 FINANCIAL_TECHNOLOGY 03:0461 (0.885); BUSINESS...
FINANCIAL_INCLUSION 03:0590 CASE_STUDY 02:0340 (0.971); BLOCKCHAIN 02:0305...
FINANCIAL_TECHNOLOGY 03:0461 FINANCIAL_SERVICES 04:0667 (0.885); BUSINESS_M...