Indexation module output assessment
All texts included in KEOPS are parsed by the Index module, and tagged with keywords. The keywords of the texts are also associated with the project to which each of them is attached.
The text corpus currently included in KEOPS involves
208
projects and contains
1887
texts coming from:
208
database project sheets 1227
documents collected by task 3.1 156
websites
KEOPS uses 5 referentials (i.e. groups of keywords) to parse the texts:
- EXPERT: includes the keywords present in the "keywords" section of the database built by WP3.
- PRETORIA: contains the keywords identified at the Pretoria workshop in 2019 (Roche et al. 2020). Elaborated from the documents provided by the participants, it highlights the relevant terms for the members of the LEAP4FNSSA project.
- AGROVOC: contains the AGROVOC thesaurus, that covers all areas of interest of the Food and Agriculture Organization (FAO) of the United Nations. AGROVOC consists of +37,000 concepts, corresponding to the prefered terms of the areas of interest, as well as synonyms. For instance, Maghreb is a synonym of the preferential term “North Africa”.
- BIOTEX: contains the most discriminative keywords (= terms) appearing in the text corpus. These keywords are identified using the software BioTex (Lossio-Ventura et al. 2016).
- EXTRACT: contains the named entities (organization, geographic location) identified and extracted from the text corpus by the software SpaCy using the module EXTRACT.
The aim of the output presented below is to evaluate the indexation methods implemented in KEOPS. Additional outputs, to evaluate KEOPS, will be provided at a later date. These outputs will serve as a basis for building the decision-making tools made available for the HLPD to drive the roadmap.
References
- Lossio-Ventura J.A., Jonquet C., Roche M., Teisseire M. 2016. Biomedical term extraction: overview and a new methodology. Information Retrieval Journal 19(1-2): 59-99.
- Helmer T., Martin P., Roche M. 2020. KEOPS - Knowledge ExtractOr Pipeline System. Montpellier. CIRAD Document, 4 p. : https://agritrop.cirad.fr/595946/
- Roche M., Helmer T., Martin P., Chaminuka P., Dimitriou I., Csorba A., Lindsten A., Lundén T., van Boheemen P., 2020. LEAP4FNSSA (WP3 - KMS): Terminology for KEOPS. doi:10.18167/DVN1/GQ8DPL, CIRAD Dataverse, V1
Mots-clés experts
P7. Number of texts and projects per PRETORIA keyword
As the PRETORIA keywords were chosen by people with knowledge of the context of the HLPD roadmap, this result has a high confidence value.
Agricultural innovation systems
Adaptation to climate change
Partnerships in agricultural research development
P8. Number of texts and projects per EXPERT keyword
As the EXPERT keywords were identified by people filling in the database project sheets , this result has a high confidence value. Only the results for the first 100 keywords, appearing most frequently in the corpus, are presented.
205
·
44
Water management
191
·
36
ecosystem services
171
·
41
nutrition security
141
·
40
sustainable agriculture
130
·
35
local communities
116
·
27
local stakeholders
99
·
29
farming practices
90
·
24
nutritional value
83
·
34
environmental impact
65
·
26
sustainable intensification
58
·
2
urban food production
55
·
22
carbon sequestration
55
·
20
climate change adaptation
54
·
21
traditional knowledge
Mots-clés Agrovoc
P9. Number of texts and projects per AGROVOC keyword mapped with PRETORIA
Agricultural innovation systems
Adaptation to climate change
Partnerships in agricultural research development
P10. Number of texts and projects per AGROVOC keyword
1065
·
132
calcium ammonium nitrate
634
·
117
farming systems
600
·
89
economic distribution
500
·
79
European Commission
P11. Average number of AGROVOC concepts per text
This indicator measures the relevance of the Agrovoc thesaurus for indexing the texts collected to process the HLPD roadmap.
Mots-clés BioTex
P12. Number of texts and projects per BIOTEX keyword
Only the results for the first 100 keywords, appearing most frequently in the corpus, are presented.
286
·
63
agricultural research
270
·
72
research institute
157
·
5
urban agriculture
131
·
15
waste management
124
·
1
tradition revisited
104
·
33
wageningen university
94
·
9
international water
88
·
2
peri-urban agriculture
81
·
7
management institute
68
·
23
international food
55
·
19
innovation partnership
45
·
10
regional economic
44
·
15
knowledge management
40
·
23
makerere university
35
·
14
eu-africa research
31
·
17
mediterranean area
29
·
12
european parliament
28
·
8
executive director
25
·
9
regional universities
25
·
9
universities forum
25
·
3
unesco-ihe institute
20
·
1
technologies revisited
19
·
5
agroforestry centre
18
·
7
world agroforestry
18
·
7
agricultural engineering
17
·
2
metropolitan district
17
·
9
economic community
17
·
10
egerton university
15
·
5
swedish university
15
·
5
climate-smart agriculture
Organisations
P13. Number of texts and projects per organization or institution of EXTRACT
Only the results for the first 100 organizations and institutions, appearing most frequently in the corpus, are presented.
283
·
62
The European Union
271
·
78
European Commission
236
·
60
the European Commission
151
·
42
the United Nations
122
·
31
Natural Resources
119
·
44
Agricultural Research
119
·
1
Urban Agriculture
116
·
36
Ministry of Agriculture
110
·
38
Sub-Saharan Africa
105
·
21
the European Union’s
P14. Number of texts and projects per organization or institution of EXPERT (field “partners” in the database)
Only the results for the first 100 organizations of institutions, appearing most frequently in the corpus, are presented.
44
Agricultural Research Centre for International Development (CIRAD)22
Research Institute for Development (IRD)19
CSIC - Spanish National Research Council15
Council for Scientific and Industrial Research (CSIR-Ghana)13
International Livestock Research Institute (ILRI)13
University of Pretoria12
International Institute of Tropical Agriculture (IITA)11
University of Copenhagen10
Cheikh Anta Diop University in Dakar (UCAD)10
French National Research Institute for Agriculture, Food and Environment (INRAE)10
University Abomey-Calavi9
Natural Resources Institute Finland (LUKE)9
Agricultural Research Council (ARC)8
Kenya Agricultural and Livestock Research Organization (KALRO)8
Sokoine University of Agriculture8
Stellenbosch University8
Eduardo Mondlane University (UEM)8
French National Institute for Agricultural Research (INRA)8
Council for Scientific and Industrial Research (CSIR)7
International Center of Insect Physiology and Ecology (ICIPE)7
National Agricultural Research Organisation (NARO)7
University of Hohenheim7
Food and Agriculture Organization of the United Nations (FAO)7
National Institute for Agricultural and Food Research and Technology (INIA)7
Hassan II Institute of Agronomy and Veterinary Sciences (IAV)7
International Center for Advanced Mediterranean Agronomic Studies (CIHEAM)7
Institut National de la Recherche Agronomique (INRA-MAROC)7
Scientific and Technological Research Council of Turkey (TÜBITAK)6
International Crops Research Institute for the Semi-Arid Tropics (ICRISAT)6
Institute of Rural Economy (IER)6
University of Natural Resources and Life Sciences, Vienna (BOKU)6
Forum for Agricultural Research in Africa (FARA)6
Technical University of Denmark (DTU)6
National Research Council, Italy (CNR)6
Institut National de la Recherche Agronomique de Tunisie (INRAT)6
Institute of Agrifood Research and Technology (IRTA)6
National Research Centre (NRC)6
German Aerospace Center (DLR)6
University of Greenwich5
University of KwaZulu-Natal5
Swedish University of Agricultural Sciences (SLU)5
International Center for Research in Agroforestry / World Agroforestry (ICRAF)5
Norwegian Institute of Bioeconomy Research (NIBIO)5
Joint Research Centre of the European Commission (JRC)5
French Agency for Food, Environmental and Occupational Health & Safety (ANSES)5
University of Antananarivo5
Ministry of Higher Education and Scientific Research5
Ministry of Education, Science and Technology (MoEST)5
National Research Foundation (NRF)5
International Water Management Institute (IWMI)5
Department of Science and Technology (DST)5
University of Ouagadougou5
International Center for Research in Agroforestry (ICRAF)5
Ghana Council for Scientific and Industrial Research (CSIR)4
International Maize and Wheat Improvement Center (CIMMYT)4
Polytechnic University of Bobo-Dioulasso4
University for Development Studies (UDS)4
Ministry of Agriculture4
Leibniz Centre for Agricultural Landscape Research (ZALF)4
Szent Istvan University4
Sahara and Sahel Observatory (OSS)4
Friedrich-Loeffler-Institut (FLI) Federal Research institut for Animal Health4
Norwegian University of Life Sciences (NMBU)4
Brazilian Agricultural Research Corporation (Embrapa)4
Félix Houphouët-Boigny University4
West and Central African Council for Agricultural Research and Development (CORAF/WECARD)4
Council for Agricultural Research and Economics (CREA)4
Agricultural University of Athens4
Algeria National Agronomic Research institute (INRAA)4
French National Research Agency (ANR)4
Federal Office for Agriculture and Food (BLE)4
Agricultural Research Centre (ARC), Egypt4
University of Jyväskylä4
Regional Universities Forum for Capacity Building in Agriculture (RUFORUM)4
Swiss Tropical and Public Health Institute (Swiss TPH)4
Institute of Tropical Medicine Antwerp3
Centre de Suivi Ecologique (CSE)3
African Forum for Agricultural Advisory Services (AFAAS)3
Federal Department of Economic Affairs, Education and Research3
University of Cape Coast P15. Comparison of P13 and P14 lists
For the 100 organizations/institutions identified in P13 and P14, are listed those exclusively present in EXTRACT, those exclusively present in EXPERT, and those present in both EXTRACT and EXPERT. Organizations and institutions are ranked from the most frequent to the least frequent
Lieux
P16. Number of texts and projects per EXTRACT location
Only the results for the first 100 locations, appearing most frequently in the corpus, are presented.
150
·
41
Sub-Saharan Africa
P17. Number of texts and projects per organization or institution of EXPERT (field “Geographic areas” in the project database)
Only the results for the first 100 locations, appearing most frequently in the corpus, are presented.
P18. Comparison of P16 and P17 lists
For the 100 locations identified in P16 and P17, are listed those exclusively present in EXTRACT, those exclusively present in EXPERT, and those present in both EXTRACT and EXPERT. Locations are ranked from the most frequent to the least frequent.
Evaluation des intersections
P19. Number of texts per EXTRACT organization and PRETORIA keyword
This output presents the thematics in which the organizations/institutions are involved. Only the results for the 100 most frequent "Institution & Thematic" pairs are presented.
432
Agriculture
·
research432
Agriculture
·
research411
Agriculture
·
management405
Development
·
research405
Development
·
research393
Development
·
management385
Agriculture
·
knowledge385
Agriculture
·
knowledge385
Agriculture
·
knowledge375
International
·
research375
International
·
research365
Agriculture
·
quality365
Development
·
knowledge365
Development
·
knowledge365
Development
·
knowledge360
Agriculture
·
capacity360
Agriculture
·
capacity358
International
·
management354
Development
·
capacity354
Development
·
capacity351
Development
·
quality337
International
·
knowledge337
International
·
knowledge337
International
·
knowledge330
International
·
quality326
International
·
impact319
University
·
research319
University
·
research318
International
·
capacity318
International
·
capacity313
Agriculture
·
implementation306
Research
·
management300
Development
·
implementation295
Agricultural
·
research295
Agricultural
·
research285
Agricultural
·
management280
University
·
management276
Institute
·
management
P20. Number of texts per EXPERT organism (field “Partners” in the project database) and PRETORIA keyword
This output presents the thematics in which the institutions are involved. Only the results for the 100 most frequent "Institution & Thematic" pairs are presented.
17
Agricultural Research Centre for International Development (CIRAD)
·
food15
Agricultural Research Centre for International Development (CIRAD)
·
research15
Agricultural Research Centre for International Development (CIRAD)
·
research13
Agricultural Research Centre for International Development (CIRAD)
·
management10
Agricultural Research Centre for International Development (CIRAD)
·
knowledge10
Agricultural Research Centre for International Development (CIRAD)
·
knowledge10
Agricultural Research Centre for International Development (CIRAD)
·
knowledge9
Agricultural Research Centre for International Development (CIRAD)
·
innovation9
Research Institute for Development (IRD)
·
research9
Research Institute for Development (IRD)
·
research8
Agricultural Research Centre for International Development (CIRAD)
·
capacity8
Agricultural Research Centre for International Development (CIRAD)
·
capacity8
Research Institute for Development (IRD)
·
food8
Makerere University
·
food8
University of Nairobi
·
food7
Makerere University
·
research7
Makerere University
·
research6
Agricultural Research Centre for International Development (CIRAD)
·
impact6
Agricultural Research Centre for International Development (CIRAD)
·
climate change6
Research Institute for Development (IRD)
·
management6
Research Institute for Development (IRD)
·
capacity6
Research Institute for Development (IRD)
·
capacity6
Makerere University
·
impact6
CSIC - Spanish National Research Council
·
food6
Council for Scientific and Industrial Research (CSIR-Ghana)
·
food6
International Livestock Research Institute (ILRI)
·
research6
International Livestock Research Institute (ILRI)
·
research6
International Institute of Tropical Agriculture (IITA)
·
food6
University Abomey-Calavi
·
research6
University Abomey-Calavi
·
research6
University of Helsinki
·
food5
Agricultural Research Centre for International Development (CIRAD)
·
implementation5
CSIC - Spanish National Research Council
·
management5
Wageningen University
·
research5
Wageningen University
·
research5
Wageningen University
·
food5
Council for Scientific and Industrial Research (CSIR-Ghana)
·
research5
Council for Scientific and Industrial Research (CSIR-Ghana)
·
research5
University of Nairobi
·
research5
University of Nairobi
·
research5
University of Nairobi
·
quality5
International Livestock Research Institute (ILRI)
·
food5
International Livestock Research Institute (ILRI)
·
knowledge5
International Livestock Research Institute (ILRI)
·
knowledge5
International Livestock Research Institute (ILRI)
·
knowledge5
International Livestock Research Institute (ILRI)
·
livestock5
University of Pretoria
·
food5
International Institute of Tropical Agriculture (IITA)
·
quality5
University of Copenhagen
·
research5
University of Copenhagen
·
research5
University Abomey-Calavi
·
food5
University Abomey-Calavi
·
knowledge5
University Abomey-Calavi
·
knowledge5
University Abomey-Calavi
·
knowledge5
University Abomey-Calavi
·
impact5
Agricultural Research Council (ARC)
·
food5
Aarhus University
·
management4
Agricultural Research Centre for International Development (CIRAD)
·
quality4
Agricultural Research Centre for International Development (CIRAD)
·
crop4
Agricultural Research Centre for International Development (CIRAD)
·
livestock4
Agricultural Research Centre for International Development (CIRAD)
·
partnership4
Research Institute for Development (IRD)
·
knowledge4
Research Institute for Development (IRD)
·
knowledge4
Research Institute for Development (IRD)
·
knowledge4
Makerere University
·
knowledge4
Makerere University
·
knowledge4
Makerere University
·
knowledge4
Makerere University
·
capacity4
Makerere University
·
capacity4
Makerere University
·
livestock4
CSIC - Spanish National Research Council
·
capacity4
CSIC - Spanish National Research Council
·
capacity4
Wageningen University
·
impact4
Council for Scientific and Industrial Research (CSIR-Ghana)
·
knowledge4
Council for Scientific and Industrial Research (CSIR-Ghana)
·
knowledge4
Council for Scientific and Industrial Research (CSIR-Ghana)
·
knowledge4
Council for Scientific and Industrial Research (CSIR-Ghana)
·
innovation4
University of Nairobi
·
capacity4
University of Nairobi
·
capacity4
International Institute of Tropical Agriculture (IITA)
·
research4
International Institute of Tropical Agriculture (IITA)
·
research4
International Institute of Tropical Agriculture (IITA)
·
management4
International Institute of Tropical Agriculture (IITA)
·
knowledge4
International Institute of Tropical Agriculture (IITA)
·
knowledge4
International Institute of Tropical Agriculture (IITA)
·
knowledge4
International Institute of Tropical Agriculture (IITA)
·
impact4
University of Copenhagen
·
quality4
University of Copenhagen
·
impact4
University Abomey-Calavi
·
management4
University Abomey-Calavi
·
capacity4
University Abomey-Calavi
·
capacity4
Natural Resources Institute Finland (LUKE)
·
food4
Natural Resources Institute Finland (LUKE)
·
capacity4
Natural Resources Institute Finland (LUKE)
·
capacity4
Kenya Agricultural and Livestock Research Organization (KALRO)
·
food4
Kenya Agricultural and Livestock Research Organization (KALRO)
·
crop4
Stellenbosch University
·
knowledge
P22a. Number of texts and projects per EXTRACT country and PRETORIA keyword
This output presents the thematics per country. All the locations identified by EXTRACT (city, regions, zones...) are brought back to the country level via Geonames. Only the results for the 100 most frequent "place & thematics" pairs are presented.
329
Netherlands
·
management324
South Africa
·
research324
South Africa
·
research318
Netherlands
·
research318
Netherlands
·
research317
South Africa
·
management309
South Africa
·
impact299
South Africa
·
knowledge299
South Africa
·
knowledge299
South Africa
·
knowledge296
Netherlands
·
quality294
Netherlands
·
knowledge294
Netherlands
·
knowledge294
Netherlands
·
knowledge285
Netherlands
·
capacity285
Netherlands
·
capacity283
South Africa
·
quality278
South Africa
·
capacity278
South Africa
·
capacity262
Kenya
·
implementation260
Ethiopia
·
management243
Netherlands
·
implementation237
South Africa
·
implementation226
Tanzania
·
management217
Ghana
·
implementation215
Uganda
·
implementation P22b. Number of texts and projects per EXTRACT location and PRETORIA keyword
This output presents the thematics per location. Only the results for the 100 most frequent "location & thematics" pairs are presented.
505
Africa
·
implementation396
Africa
·
climate change329
Netherlands
·
management324
South Africa
·
research324
South Africa
·
research318
Netherlands
·
research318
Netherlands
·
research317
South Africa
·
management314
Europe
·
implementation309
South Africa
·
impact299
South Africa
·
knowledge299
South Africa
·
knowledge299
South Africa
·
knowledge296
Netherlands
·
quality294
Netherlands
·
knowledge294
Netherlands
·
knowledge294
Netherlands
·
knowledge285
Netherlands
·
capacity285
Netherlands
·
capacity283
South Africa
·
quality278
South Africa
·
capacity278
South Africa
·
capacity278
Programme
·
management P23. Number of texts and projects per EXPERT location (field “Geographic zone” in the project database) and PRETORIA keyword
This output presents the thematics per location. Only the results for the 100 most frequent "location & thematics" pairs are presented.
521
Africa
·
implementation408
Africa
·
climate change364
Netherlands
·
management353
Netherlands
·
research353
Netherlands
·
research329
Netherlands
·
knowledge329
Netherlands
·
knowledge329
Netherlands
·
knowledge324
Netherlands
·
quality323
South Africa
·
research323
South Africa
·
research321
Europe
·
implementation315
South Africa
·
management313
Netherlands
·
capacity313
Netherlands
·
capacity308
South Africa
·
impact297
South Africa
·
knowledge297
South Africa
·
knowledge297
South Africa
·
knowledge283
Africa
·
food production281
South Africa
·
quality278
South Africa
·
capacity278
South Africa
·
capacity265
Netherlands
·
implementation263
Kenya
·
implementation