5.2. Método de revisão Impresso#
Esse notebook[1] foi desenvolvido por Ehrmann, Maud, Estelle Bunout, e Marten Düring e apresenta um método de avaliação de interfaces gráficas de jornais históricos. Os resultados da pesquisa foram publicados 2017 no artigo Historical Newspaper User Interfaces: A Review.
Busco aqui aplicar o mesmo método a partir dos dados das três hemerotecas estudadas: Hemeroteca Digital da Biblioteca Nacional, Hemeroteca Digital da Biblioteca Nacional de Portugal, Hemeroteca Municipal de Lisboa.
Os gráficos gerados estão armazenados em charts/.
Nota
Esse notebook compõe o repositórioSurvey of Digitized Newspaper Interfaces (dataset and Notebooks) e está licenciado sob a licença CC BY-NC-SA 4.0.
Busquei manter o código e documentação mais próximo do original. Quando foi necessário realizar mudanças, elas foram documentadas.
%matplotlib inline
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from radar_factory import build_single_radar, build_multiple_radar, build_single_radar_free
Preparing data#
Loading the review spreasheet as a multi-index (tsv)#
base = pd.read_csv("../../../data/avaliacao_interfaces/InterfaceReview-pt-2023.csv", sep="\t", index_col= [0,1], skipinitialspace=True)
base.head()
| Hemeroteca Digital Brasileira (HDB) | Biblioteca Nacional Digital de Portugal (BND-PT) | Hemeroteca Digital de Lisboa (HDL) | ||
|---|---|---|---|---|
| Property family | Property | |||
| interface | URL | http://memoria.bn.br/hdb/ | https://bndigital.bnportugal.gov.pt/ | https://hemerotecadigital.cm-lisboa.pt/ |
| Target area | BR | PT | PT | |
| Creator | Fundação Biblioteca Nacional | Biblioteca Nacional de Portugal | Bibliotecas de Lisboa | |
| Purpose and scope | Coleção de periódicos digitalizados do acervo ... | Acervo digitalizado da Biblioteca Nacional de ... | Periódicos, legislação obras raras digitalizad... | |
| Approximate date of creation | 2012 | u | u |
Testing some indexing#
# about multilingual collections
base.loc['newspaper collection', 'Languages of the collections'].value_counts()
pt 3
Name: (newspaper collection, Languages of the collections), dtype: int64
# about multilingual collections
base.loc['interface', 'Access model'].value_counts()
free 3
Name: (interface, Access model), dtype: int64
# about multilingual collections
base.loc['interface', 'Interface provider'].value_counts()
Instituição Pública 2
Private (DocPro) 1
Name: (interface, Interface provider), dtype: int64
Some cleaning and preparation#
def trim_all_columns(df):
"""
Trim whitespace from ends of each value across all series in dataframe
"""
trim_strings = lambda x: x.strip() if type(x) is str else x
return df.applymap(trim_strings)
# trim
base = trim_all_columns(base)
# removing non binary features
base = base.drop('interface', level=0)
base = base.drop('newspaper collection', level=0)
base = base.drop('Languages of the collections', level=1) # from 'newspaper collection'
base = base.drop('Other', level=1)
base = base.drop('Download options (file formats)', level=1)
#Replace n and y by 0 and 1
base = base.replace(to_replace=['y', 'y?', 'y (annotations)', 'y (requires user account - free)', 'n', '?', 'u', 'n?', 'n (?)', 'n (but can signal mistakes)'],
value=[1,1,1,1,0,0,0,0,0,0])
base.head()
| Hemeroteca Digital Brasileira (HDB) | Biblioteca Nacional Digital de Portugal (BND-PT) | Hemeroteca Digital de Lisboa (HDL) | ||
|---|---|---|---|---|
| Property family | Property | |||
| newspaper metadata | Alternative titles | 0 | 1 | 1 |
| Place of publication | 1 | 1 | 1 | |
| Geographic coverage | 1 | 0 | 0 | |
| Publisher | 0 | 1 | 1 | |
| Date range | 1 | 1 | 1 |
Interface charts#
Generic counts#
base.loc[:,'Total'] = base.sum(axis=1, numeric_only=True)
base['Total (%)'] = base['Total']/base['Total'].sum() * 100
base.head()
| Hemeroteca Digital Brasileira (HDB) | Biblioteca Nacional Digital de Portugal (BND-PT) | Hemeroteca Digital de Lisboa (HDL) | Total | Total (%) | ||
|---|---|---|---|---|---|---|
| Property family | Property | |||||
| newspaper metadata | Alternative titles | 0 | 1 | 1 | 2 | 2.127660 |
| Place of publication | 1 | 1 | 1 | 3 | 3.191489 | |
| Geographic coverage | 1 | 0 | 0 | 1 | 1.063830 | |
| Publisher | 0 | 1 | 1 | 2 | 2.127660 | |
| Date range | 1 | 1 | 1 | 3 | 3.191489 |
base.shape
(125, 5)
zeros = base.loc[base['Total'].values == 0]
all = base.loc[base['Total'].values == 24]
upper = base.loc[base['Total'].values >= 12]
three_quarter = base.loc[base['Total'].values >= 18]
lower = base.loc[ (base['Total'].values <12) & (base['Total'].values != 0)]
lower_full = base.loc[base['Total'].values <12]
zeros
| Hemeroteca Digital Brasileira (HDB) | Biblioteca Nacional Digital de Portugal (BND-PT) | Hemeroteca Digital de Lisboa (HDL) | Total | Total (%) | ||
|---|---|---|---|---|---|---|
| Property family | Property | |||||
| newspaper metadata | ISSN, OCLC, LCCN | 0 | 0 | 0 | 0 | 0.0 |
| browsing | By user tag | 0 | 0 | 0 | 0 | 0.0 |
| search | Fuzzy search | 0 | 0 | 0 | 0 | 0.0 |
| Proximity search | 0 | 0 | 0 | 0 | 0.0 | |
| Limit by NP segments | 0 | 0 | 0 | 0 | 0.0 | |
| ... | ... | ... | ... | ... | ... | ... |
| connectivity | SW technologies | 0 | 0 | 0 | 0 | 0.0 |
| apis | Link to source code of the interface | 0 | 0 | 0 | 0 | 0.0 |
| API | 0 | 0 | 0 | 0 | 0.0 | |
| IIIF Image API | 0 | 0 | 0 | 0 | 0.0 | |
| IIIF Presentation API | 0 | 0 | 0 | 0 | 0.0 |
73 rows × 5 columns
all
| Hemeroteca Digital Brasileira (HDB) | Biblioteca Nacional Digital de Portugal (BND-PT) | Hemeroteca Digital de Lisboa (HDL) | Total | Total (%) | ||
|---|---|---|---|---|---|---|
| Property family | Property |
lower.shape
(52, 5)
upper.shape
(0, 5)
lower_full.shape
(125, 5)
three_quarter.shape
(0, 5)
Group by for aggregated counts#
# Sum values of level 1 => 'grade' of each interface per family of features
level_0 = base.groupby(level=0).sum()
#level_0 = base.groupby(level=0).sum(numeric_only=True)
# re-order index
level_0 = level_0.reindex(["newspaper metadata",
"apis",
"connectivity",
"info on digitization",
"enrichment",
"user interaction",
"viewer",
"result display",
"result filtering",
"result sorting",
"search",
"browsing"
])
# copy level_0 to analyze
level_0_copy = level_0.copy()
level_0_copy
| Hemeroteca Digital Brasileira (HDB) | Biblioteca Nacional Digital de Portugal (BND-PT) | Hemeroteca Digital de Lisboa (HDL) | Total | Total (%) | |
|---|---|---|---|---|---|
| Property family | |||||
| newspaper metadata | 4 | 9 | 8 | 21 | 22.340426 |
| apis | 0 | 0 | 0 | 0 | 0.000000 |
| connectivity | 0 | 1 | 0 | 1 | 1.063830 |
| info on digitization | 1 | 1 | 0 | 2 | 2.127660 |
| enrichment | 0 | 0 | 0 | 0 | 0.000000 |
| user interaction | 1 | 1 | 0 | 2 | 2.127660 |
| viewer | 6 | 8 | 4 | 18 | 19.148936 |
| result display | 3 | 3 | 1 | 7 | 7.446809 |
| result filtering | 3 | 6 | 0 | 9 | 9.574468 |
| result sorting | 2 | 4 | 0 | 6 | 6.382979 |
| search | 6 | 11 | 0 | 17 | 18.085106 |
| browsing | 3 | 4 | 4 | 11 | 11.702128 |
Observations per feature family#
# Não é necessário pelo pequeno número de interfaces analisadas. Vamos trabalhar com o valor absoluto
#Total sum per row, ie. per feature family (=> how good are all interfaces for a certain aspect):
#level_0.loc[:,'Total'] = level_0.sum(axis=1)
#level_0['Total (%)'] = level_0['Total']/level_0['Total'].sum() * 100
# getting the labels
labels = level_0_copy.index
Test radar with one interface#
# Get all rows, just for the first columns, transpose it (for the radar factory) and take only the values
first_interface = level_0_copy.iloc[:12, :1].T.values
build_single_radar(labels, values=first_interface, title=level_0_copy.columns[0], grid=[2,4,6,8], figure_title='HDB', output_type='png')
Radar view for each interface in one figure#
:warning: Vamos trabalhar com valores inteiros pelo pequeno número de interfaces analisadas. Assim as próximas células não serão executadas. Utilizaremos o df level_0_copy para as próximas análises.
# same with percentages
#level_0_percent = base.groupby(level=0).sum().apply(lambda x: 100*x/float(x.sum()))
# re-order index
#level_0_percent = level_0_percent.reindex(["newspaper metadata",
# "apis",
# "connectivity",
# "info on digitization",
# "enrichment",
# "user interaction",
# "viewer",
# "result display",
# "result filtering",
# "result sorting",
# "search",
# "browsing"
# ])
#level_0_percent.iloc[:12, 4].index
# checking we have 100 everywhere
#level_0_percent.sum()
#level_0_percent.max().max()
#level_0_percent.iloc[:12, :24]
all_interfaces_percents = level_0_copy.iloc[:12, :].T.values
build_multiple_radar(labels, all_interfaces_percents, level_0_copy.columns[:], 'all-interfaces-single', output_type='png')
Global radar view (i.e. for all interfaces) over all features#
level_0_copy.iloc[:,4]
Property family
newspaper metadata 22.340426
apis 0.000000
connectivity 1.063830
info on digitization 2.127660
enrichment 0.000000
user interaction 2.127660
viewer 19.148936
result display 7.446809
result filtering 9.574468
result sorting 6.382979
search 18.085106
browsing 11.702128
Name: Total (%), dtype: float64
# take only the last column: Total per feature family, in percent
values = level_0_copy.iloc[:,4:].T.values
build_single_radar_free(labels, values, "All interfaces",figure_title='all-interfaces-global', output_type='png')
Metadata (global)#
metadata = base.loc['newspaper metadata'].copy()
metadata.loc[:,'Total'] = metadata.sum(axis=1)
#metadata['Total (%)'] = metadata['Total']/metadata['Total'].sum() * 100
metadata['Total (%)'] = metadata['Total']/24 * 100
metadata.loc[:,'Total']
Property
Alternative titles 6.127660
Place of publication 9.191489
Geographic coverage 3.063830
Publisher 6.127660
Date range 9.191489
Periodicity 3.063830
ISSN, OCLC, LCCN 0.000000
External links 6.127660
Historical description 3.063830
Language 3.063830
Calendar view of issues 6.127660
Indication of archive holder 9.191489
Name: Total, dtype: float64
metadata.shape[0]
12
metadata['Total']
Property
Alternative titles 6.127660
Place of publication 9.191489
Geographic coverage 3.063830
Publisher 6.127660
Date range 9.191489
Periodicity 3.063830
ISSN, OCLC, LCCN 0.000000
External links 6.127660
Historical description 3.063830
Language 3.063830
Calendar view of issues 6.127660
Indication of archive holder 9.191489
Name: Total, dtype: float64
metadata.sort_values('Total (%)').loc[:, 'Total (%)']
Property
ISSN, OCLC, LCCN 0.000000
Geographic coverage 12.765957
Periodicity 12.765957
Historical description 12.765957
Language 12.765957
Alternative titles 25.531915
Publisher 25.531915
External links 25.531915
Calendar view of issues 25.531915
Place of publication 38.297872
Date range 38.297872
Indication of archive holder 38.297872
Name: Total (%), dtype: float64
values_metadata = metadata.iloc[:,4:].T.values
build_single_radar_free(metadata.index, values_metadata, title="Newspaper metadata", figure_title="metadata-global", output_type='png')
Browsing#
browsing = base.loc['browsing'].copy()
browsing.loc[:,'Total'] = browsing.sum(axis=1)
browsing['Total (%)'] = browsing['Total']/24 * 100
browsing.loc[:,'Total']
Property
By date 9.191489
By title 9.191489
By place of publication 9.191489
By user tag 0.000000
By newspaper thematic (metadata) 6.127660
Name: Total, dtype: float64
browsing.shape[0]
5
browsing.sort_values('Total (%)').loc[:, 'Total (%)']
Property
By user tag 0.000000
By newspaper thematic (metadata) 25.531915
By date 38.297872
By title 38.297872
By place of publication 38.297872
Name: Total (%), dtype: float64
values_browsing = browsing.iloc[:,4:].T.values
build_single_radar_free(browsing.index, values_browsing, title="Browsing", figure_title="browsing-global", output_type='png')
Search#
search = base.loc['search'].copy()
search.loc[:,'Total'] = search.sum(axis=1)
search['Total (%)'] = search['Total']/24 * 100
search['Total']
Property
Keyword search 6.12766
Query autocomplete 3.06383
Boolean operators 3.06383
Phrase search 6.12766
Fuzzy search 0.00000
Wild card 6.12766
Proximity search 0.00000
Limit by date range 6.12766
Limit by language 3.06383
Limit by NP title(s) 6.12766
Limit by place of publication 6.12766
Limit by NP thematic (from metadata) 3.06383
Limit by NP segments 0.00000
Limit by article category 0.00000
Limit by article length 0.00000
Limit by archival holder 0.00000
Limit by license 3.06383
Query suggestion 0.00000
Search by NE 0.00000
Name: Total, dtype: float64
search.shape[0]
19
search.sort_values('Total (%)').loc[:, 'Total (%)']
Property
Search by NE 0.000000
Fuzzy search 0.000000
Limit by archival holder 0.000000
Proximity search 0.000000
Limit by article length 0.000000
Query suggestion 0.000000
Limit by article category 0.000000
Limit by NP segments 0.000000
Query autocomplete 12.765957
Boolean operators 12.765957
Limit by license 12.765957
Limit by language 12.765957
Limit by NP thematic (from metadata) 12.765957
Keyword search 25.531915
Limit by date range 25.531915
Wild card 25.531915
Phrase search 25.531915
Limit by place of publication 25.531915
Limit by NP title(s) 25.531915
Name: Total (%), dtype: float64
values_search = search.iloc[:,4:].T.values
build_single_radar_free(search.index, values_search, title="Search", figure_title="search-global", output_type='png')
Result display#
rd = base.loc['result display'].copy()
rd.loc[:,'Total'] = rd.sum(axis=1)
rd['Total (%)'] = rd['Total']/24 * 100
rd['Total']
Property
Distribution over time 6.127660
Distribution by publication place 0.000000
Distribution by NP 9.191489
Distribution by place names in articles 0.000000
Snippet preview 3.063830
Search highlight in facsimiles 3.063830
Search highlight in text 0.000000
Ngrams 0.000000
Name: Total, dtype: float64
rd.shape[0]
8
rd.sort_values('Total (%)').loc[:, 'Total (%)']
Property
Distribution by publication place 0.000000
Distribution by place names in articles 0.000000
Search highlight in text 0.000000
Ngrams 0.000000
Snippet preview 12.765957
Search highlight in facsimiles 12.765957
Distribution over time 25.531915
Distribution by NP 38.297872
Name: Total (%), dtype: float64
values_rd = rd.iloc[:,4:].T.values
build_single_radar_free(rd.index, values_rd, title="Result display", figure_title="result-display-global", output_type='png')
Result filtering#
rf = base.loc['result filtering'].copy()
rf.loc[:,'Total'] = rf.sum(axis=1)
rf['Total (%)'] = rf['Total']/24 * 100
rf['Total']
Property
By NP titles 3.06383
By periodicity 0.00000
By NP orientation 0.00000
By newspaper thematic (metadata) 3.06383
By content types 3.06383
By sections 0.00000
By events 0.00000
By persons 0.00000
By organisations 0.00000
By places mentioned in text 0.00000
By time period 6.12766
By topics 0.00000
By manual tags 0.00000
By publication place 6.12766
By archive 0.00000
By publisher 0.00000
By article length 0.00000
By authors 0.00000
By segmentation level 0.00000
By language 3.06383
By license 3.06383
By online pub. date 0.00000
Name: Total, dtype: float64
rf.shape[0]
22
rf.sort_values('Total (%)').loc[:, 'Total (%)']
Property
By online pub. date 0.000000
By periodicity 0.000000
By NP orientation 0.000000
By segmentation level 0.000000
By sections 0.000000
By events 0.000000
By persons 0.000000
By organisations 0.000000
By places mentioned in text 0.000000
By authors 0.000000
By topics 0.000000
By manual tags 0.000000
By article length 0.000000
By archive 0.000000
By publisher 0.000000
By language 12.765957
By NP titles 12.765957
By license 12.765957
By content types 12.765957
By newspaper thematic (metadata) 12.765957
By publication place 25.531915
By time period 25.531915
Name: Total (%), dtype: float64
values_rf = rf.iloc[:,4:].T.values
build_single_radar_free(rf.index, values_rf, title="Result filtering", figure_title="result-filtering-global", output_type='png')
Result sorting#
rs = base.loc['result sorting'].copy()
rs.loc[:,'Total'] = rs.sum(axis=1)
rs['Total (%)'] = rs['Total']/24 * 100
rs['Total']
Property
By relevance 3.06383
By date 6.12766
By NP title 6.12766
By article title 0.00000
By content type 0.00000
By online pub. date 3.06383
By author 0.00000
By quality of text 0.00000
By language 0.00000
By popularity 0.00000
Name: Total, dtype: float64
rs.shape[0]
10
rs.sort_values('Total (%)').loc[:, 'Total (%)']
Property
By article title 0.000000
By content type 0.000000
By author 0.000000
By quality of text 0.000000
By language 0.000000
By popularity 0.000000
By relevance 12.765957
By online pub. date 12.765957
By date 25.531915
By NP title 25.531915
Name: Total (%), dtype: float64
values_rs = rs.iloc[:,4:].T.values
build_single_radar_free(rs.index, values_rs, title="Result sorting", figure_title="result-sorting-global", output_type='png')
Viewer#
viewer = base.loc['viewer'].copy()
viewer.loc[:,'Total'] = viewer.sum(axis=1)
viewer['Total (%)'] = viewer['Total']/24 * 100
viewer.loc[:,'Total']
Property
Facsimile displayed 9.191489
OCRed text display 3.063830
Show full page 9.191489
Interactive mini-map 3.063830
Overview of avail. issues 9.191489
Search in viewed page 6.127660
Option to continue to next page 9.191489
Option to continue to next result 6.127660
Name: Total, dtype: float64
viewer.shape[0]
8
viewer.sort_values('Total (%)').loc[:, 'Total (%)']
Property
OCRed text display 12.765957
Interactive mini-map 12.765957
Search in viewed page 25.531915
Option to continue to next result 25.531915
Facsimile displayed 38.297872
Show full page 38.297872
Overview of avail. issues 38.297872
Option to continue to next page 38.297872
Name: Total (%), dtype: float64
values_viewer = viewer.iloc[:,4:].T.values
build_single_radar_free(viewer.index, values_viewer, title="Viewer", figure_title="viewer-global", output_type='png')
Info on digitization#
info = base.loc['info on digitization'].copy()
info.loc[:,'Total'] = info.sum(axis=1)
info['Total (%)'] = info['Total']/24 * 100
info.loc[:,'Total']
Property
OLR at article level 0.00000
OCR confidence scores 0.00000
OLR confidence scores 0.00000
Documentation of biases 0.00000
Search result relevance score 0.00000
Digitisation date at title level 0.00000
Scan resolution 0.00000
Used OCR tools 0.00000
Copyright notice 6.12766
Documentation of scan methods 0.00000
Name: Total, dtype: float64
info.shape[0]
10
info.sort_values('Total (%)').loc[:, 'Total (%)']
Property
OLR at article level 0.000000
OCR confidence scores 0.000000
OLR confidence scores 0.000000
Documentation of biases 0.000000
Search result relevance score 0.000000
Digitisation date at title level 0.000000
Scan resolution 0.000000
Used OCR tools 0.000000
Documentation of scan methods 0.000000
Copyright notice 25.531915
Name: Total (%), dtype: float64
values_info = info.iloc[:,4:].T.values
build_single_radar_free(info.index, values_info, title="Information on digitization", figure_title="info-global", output_type='png')
User interaction#
user = base.loc['user interaction'].copy()
user.loc[:,'Total'] = user.sum(axis=1)
user['Total (%)'] = user['Total']/24 * 100
user['Total']
Property
Save articles to favorites 0.00000
Save queries to favorites 0.00000
Tag articles 0.00000
Keep track of viewed materials 0.00000
Article recommendations 0.00000
Permalinks 6.12766
Export citation 0.00000
Option to correct OCR 0.00000
Option to correct OLR 0.00000
Add/edit of metadata 0.00000
Screenshot tool 0.00000
Bulk downloads 0.00000
Organise articles in collections 0.00000
Contrastive view of collections 0.00000
Name: Total, dtype: float64
user.shape[0]
14
user.sort_values('Total (%)').loc[:, 'Total (%)']
Property
Save articles to favorites 0.000000
Save queries to favorites 0.000000
Tag articles 0.000000
Keep track of viewed materials 0.000000
Article recommendations 0.000000
Export citation 0.000000
Option to correct OCR 0.000000
Option to correct OLR 0.000000
Add/edit of metadata 0.000000
Screenshot tool 0.000000
Bulk downloads 0.000000
Organise articles in collections 0.000000
Contrastive view of collections 0.000000
Permalinks 25.531915
Name: Total (%), dtype: float64
values_user = user.iloc[:,4:].T.values
build_single_radar_free(user.index, values_user, title="User interaction", figure_title="user-global", output_type='png')
Enrichment#
enrich = base.loc['enrichment'].copy()
enrich = enrich.drop('Query')
enrich.loc[:,'Total'] = enrich.sum(axis=1)
enrich['Total (%)'] = enrich['Total']/24 * 100
enrich['Total']
/tmp/ipykernel_52652/2409441392.py:3: FutureWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`
enrich.loc[:,'Total'] = enrich.sum(axis=1)
Property
NERC 0.0
Entity linking 0.0
Automatic post-OCR correction 0.0
Crowd-sourced post-OCR correction 0.0
Topic Modeling 0.0
Text re-use 0.0
Sentiment Analysis 0.0
Recommendations 0.0
Event detection 0.0
Name: Total, dtype: float64
enrich.sort_values('Total (%)').loc[:, 'Total (%)']
Property
NERC 0.0
Entity linking 0.0
Automatic post-OCR correction 0.0
Crowd-sourced post-OCR correction 0.0
Topic Modeling 0.0
Text re-use 0.0
Sentiment Analysis 0.0
Recommendations 0.0
Event detection 0.0
Name: Total (%), dtype: float64
enrich.iloc[:,4:]
| Total (%) | |
|---|---|
| Property | |
| NERC | 0.0 |
| Entity linking | 0.0 |
| Automatic post-OCR correction | 0.0 |
| Crowd-sourced post-OCR correction | 0.0 |
| Topic Modeling | 0.0 |
| Text re-use | 0.0 |
| Sentiment Analysis | 0.0 |
| Recommendations | 0.0 |
| Event detection | 0.0 |
values_enrich = enrich.iloc[:,4:].T.values
build_single_radar_free(enrich.index, values_enrich, title="Enrichment", figure_title="enrichment-global", output_type='png')
Connectivity#
conn = base.loc['connectivity'].copy()
conn.loc[:,'Total'] = conn.sum(axis=1)
conn['Total (%)'] = conn['Total']/24 * 100
conn['Total']
Property
Third party identifiers 3.06383
Links to other repositories 0.00000
SW technologies 0.00000
Name: Total, dtype: float64
conn.shape[0]
3
conn.sort_values('Total (%)').loc[:, 'Total (%)']
Property
Links to other repositories 0.000000
SW technologies 0.000000
Third party identifiers 12.765957
Name: Total (%), dtype: float64
values_conn = conn.iloc[:,4:].T.values
build_single_radar_free(conn.index, values_conn, title="Connectivity", figure_title="connectivity-global", output_type='png')
APIs#
apis = base.loc['apis'].copy()
apis.loc[:,'Total'] = apis.sum(axis=1)
apis['Total (%)'] = apis['Total']/24 * 100
apis['Total']
/tmp/ipykernel_52652/1500865402.py:2: FutureWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`
apis.loc[:,'Total'] = apis.sum(axis=1)
Property
Link to source code of the interface 0.0
API 0.0
IIIF Image API 0.0
IIIF Presentation API 0.0
Name: Total, dtype: float64
apis.shape[0]
4
apis.sort_values('Total (%)').loc[:, 'Total (%)']
Property
Link to source code of the interface 0.0
API 0.0
IIIF Image API 0.0
IIIF Presentation API 0.0
Name: Total (%), dtype: float64
values_apis = apis.iloc[:,4:].T.values
build_single_radar_free(apis.index, values_apis, title="APIs", figure_title="apis-global", output_type='png')
Mapping#
:warning: Ainda falta criar esse tsv para as hemerotecas em protuguês.
base_mapping = pd.read_csv("../../../data/avaliacao_interfaces/InterfaceReview-pt-2023-mapping.tsv", sep="\t", index_col= [0,1,2], skipinitialspace=True)
base_mapping.head()
| Hemeroteca Digital Brasileira (HDB) | Biblioteca Nacional Digital de Portugal (BND-PT) | Hemeroteca Digital de Lisboa (HDL) | |||
|---|---|---|---|---|---|
| High level Criteria | Property family | Property | |||
| Interfaces | interface | URL | http://memoria.bn.br/hdb/ | https://bndigital.bnportugal.gov.pt/ | https://hemerotecadigital.cm-lisboa.pt/ |
| Target area | BR | PT | PT | ||
| Creator | Fundação Biblioteca Nacional | Biblioteca Nacional de Portugal | Bibliotecas de Lisboa | ||
| Purpose and scope | Coleção de periódicos digitalizados do acervo ... | Acervo digitalizado da Biblioteca Nacional de ... | Periódicos, legislação obras raras digitalizad... | ||
| Approximate date of creation | 2012 | u | u |
# trim
base_mapping = trim_all_columns(base_mapping)
base_mapping = base_mapping.drop('Interfaces', level=0)
base_mapping = base_mapping.drop('Other', level=2)
base_mapping = base_mapping.drop('Download options (file formats)', level=2)
base_mapping = base_mapping.drop('Newspaper date range', level=2)
base_mapping = base_mapping.drop('Number of newspaper titles', level=2)
#Replace n and y by 0 and 1 (ideally to clean in spreadsheet)
base_mapping = base_mapping.replace(to_replace=['y', 'y?', 'y (annotations)', 'y (requires user account - free)', 'n', '?', 'u', 'n?', 'n (?)', 'n (but can signal mistakes)'],
value=[1,1,1,1,0,0,0,0,0,0])
mapping = base_mapping.groupby(level=0).sum(numeric_only=True)
mapping.loc[:,'Total'] = mapping.sum(axis=1)
mapping['Total (%)'] = mapping['Total']/mapping['Total'].sum() * 100
values = mapping.iloc[:,4]
values
High level Criteria
Connectivity 3.092784
Content filtering 15.463918
Content search 32.989691
Generosity 9.278351
Source criticism 37.113402
User content management/exploration 2.061856
Name: Total (%), dtype: float64
values = mapping.iloc[:,4:].T.values
build_single_radar_free(mapping.index, values, "High-level criteria",figure_title='high-level-criteria',output_type='png')
