Co-occurrence Matrix#
>>> from techminer2.tech_mining.co_occurrence import co_occurrence_matrix
>>> matrix = co_occurrence_matrix(
... #
... # FUNCTION PARAMS:
... columns='author_keywords',
... rows='authors',
... #
... # COLUMN PARAMS:
... col_top_n=None,
... col_occ_range=(2, None),
... col_gc_range=(None, None),
... col_custom_items=None,
... #
... # ROW PARAMS:
... row_top_n=None,
... row_occ_range=(2, None),
... row_gc_range=(None, None),
... row_custom_items=None,
... #
... # DATABASE PARAMS:
... root_dir="example/",
... database="main",
... year_filter=(None, None),
... cited_by_filter=(None, None),
... )
>>> matrix.df_
author_keywords FINTECH 31:5168 ... P2P_LENDING 02:0161
authors ...
Jagtiani J. 3:0317 3 ... 2
Gomber P. 2:1065 1 ... 0
Hornuf L. 2:0358 2 ... 0
Gai K. 2:0323 2 ... 0
Qiu M. 2:0323 2 ... 0
Sun X./3 2:0323 2 ... 0
Lemieux C. 2:0253 2 ... 1
Dolata M. 2:0181 2 ... 0
Schwabe G. 2:0181 2 ... 0
Zavolokina L. 2:0181 2 ... 0
[10 rows x 12 columns]
>>> matrix.heat_map_
<pandas.io.formats.style.Styler object ...
>>> matrix.list_cells_.head()
row column matrix_value
0 FINTECH 31:5168 Jagtiani J. 3:0317 3
1 FINTECH 31:5168 Gomber P. 2:1065 1
2 FINTECH 31:5168 Hornuf L. 2:0358 2
3 FINTECH 31:5168 Gai K. 2:0323 2
4 FINTECH 31:5168 Qiu M. 2:0323 2
>>> print(matrix.prompt_)
Your task is ...
>>> matrix = co_occurrence_matrix(
... #
... # FUNCTION PARAMS:
... columns='author_keywords',
... rows=None,
... #
... # COLUMN PARAMS:
... col_top_n=10,
... col_occ_range=(None, None),
... col_gc_range=(None, None),
... col_custom_items=None,
... #
... # ROW PARAMS:
... row_top_n=None,
... row_occ_range=(2, None),
... row_gc_range=(None, None),
... row_custom_items=None,
... #
... # DATABASE PARAMS:
... root_dir="example/",
... database="main",
... year_filter=(None, None),
... cited_by_filter=(None, None),
... )
>>> matrix.df_
author_keywords FINTECH 31:5168 ... CASE_STUDY 02:0340
author_keywords ...
FINTECH 31:5168 31 ... 2
INNOVATION 07:0911 5 ... 0
FINANCIAL_SERVICES 04:0667 3 ... 0
FINANCIAL_INCLUSION 03:0590 3 ... 1
FINANCIAL_TECHNOLOGY 03:0461 2 ... 0
CROWDFUNDING 03:0335 2 ... 0
MARKETPLACE_LENDING 03:0317 3 ... 0
BUSINESS_MODELS 02:0759 2 ... 0
CYBER_SECURITY 02:0342 2 ... 0
CASE_STUDY 02:0340 2 ... 2
[10 rows x 10 columns]
>>> matrix.list_cells_.head()
row column matrix_value
0 FINTECH 31:5168 FINTECH 31:5168 31
1 FINTECH 31:5168 INNOVATION 07:0911 5
2 FINTECH 31:5168 FINANCIAL_SERVICES 04:0667 3
3 FINTECH 31:5168 FINANCIAL_INCLUSION 03:0590 3
4 FINTECH 31:5168 FINANCIAL_TECHNOLOGY 03:0461 2
>>> print(matrix.prompt_)
Your task is ...