Changes

              
    

          
          
        
        

            
f
1
{
f
1
{
            
2
  "acknowledgement": "The authors gratefully acknowledge the financial 
2
  "acknowledgement": "The authors gratefully acknowledge the financial 
            
3
support for this research by the following organizations and agencies: 
3
support for this research by the following organizations and agencies: 
            
4
Kenya Education Network Trust (KENET) - Grant Number AGMT1170; 
4
Kenya Education Network Trust (KENET) - Grant Number AGMT1170; 
            
5
Foreign, Commonwealth & Development Office (FCDO) [IMC-Grant 21108]; 
5
Foreign, Commonwealth & Development Office (FCDO) [IMC-Grant 21108]; 
            
6
Australian Centre for International Agricultural Research (ACIAR) 
6
Australian Centre for International Agricultural Research (ACIAR) 
            
7
(ProteinAfrica \u2013 Grant No: LS/2020/154), the Rockefeller 
7
(ProteinAfrica \u2013 Grant No: LS/2020/154), the Rockefeller 
            
8
Foundation (WAVE-IN\u2014 Grant No: 2021 FOD 030); Bill & Melinda 
8
Foundation (WAVE-IN\u2014 Grant No: 2021 FOD 030); Bill & Melinda 
            
9
Gates Foundation (INV-032416); IKEA Foundation (G-2204-02144), Horizon 
9
Gates Foundation (INV-032416); IKEA Foundation (G-2204-02144), Horizon 
            
10
Europe (NESTLER - Project: 101060762 - HORIZON-CL6-2021- 
10
Europe (NESTLER - Project: 101060762 - HORIZON-CL6-2021- 
            
11
FARM2FORK-01), the Curt Bergfors Foundation Food Planet Prize Award; 
11
FARM2FORK-01), the Curt Bergfors Foundation Food Planet Prize Award; 
            
12
Norwegian Agency for Development Cooperation, the Section for 
12
Norwegian Agency for Development Cooperation, the Section for 
            
13
Research, Innovation and Higher Education grant number RAF\u20133058 
13
Research, Innovation and Higher Education grant number RAF\u20133058 
            
14
KEN\u201318/0005 (CAP\u2013Africa); the Swedish International 
14
KEN\u201318/0005 (CAP\u2013Africa); the Swedish International 
            
15
Development Cooperation Agency (SIDA); the Swiss Agency for 
15
Development Cooperation Agency (SIDA); the Swiss Agency for 
            
16
Development and Cooperation (SDC); the Australian Centre for 
16
Development and Cooperation (SDC); the Australian Centre for 
            
17
International Agricultural Research (ACIAR); the Norwegian Agency for 
17
International Agricultural Research (ACIAR); the Norwegian Agency for 
            
18
Development Cooperation (NORAD); the German Federal Ministry for 
18
Development Cooperation (NORAD); the German Federal Ministry for 
            
19
Economic Cooperation and Development (BMZ); the Federal Democratic 
19
Economic Cooperation and Development (BMZ); the Federal Democratic 
            
20
Republic of Ethiopia; and the Government of the Republic of Kenya. The 
20
Republic of Ethiopia; and the Government of the Republic of Kenya. The 
            
21
funders had no role in study design, data collection and analysis, 
21
funders had no role in study design, data collection and analysis, 
            
22
decision to publish, or preparation of the manuscript. ",
22
decision to publish, or preparation of the manuscript. ",
            
23
  "administrative_areas": "Nairobi",
23
  "administrative_areas": "Nairobi",
            
24
  "author": null,
24
  "author": null,
            
25
  "author_email": null,
25
  "author_email": null,
            
26
  "citation_narrative": "DOI: 10.3389/frai.2024.1403593",
26
  "citation_narrative": "DOI: 10.3389/frai.2024.1403593",
            
27
  "collaborators": "[{\"collaborator\": \"Henry Kyalo\"}, 
27
  "collaborators": "[{\"collaborator\": \"Henry Kyalo\"}, 
            
28
{\"collaborator\": \"Henri E. Z. Tonnang\"}, {\"collaborator\": 
28
{\"collaborator\": \"Henri E. Z. Tonnang\"}, {\"collaborator\": 
            
29
\"James Egonyu\"}, {\"collaborator\": \"John Olukuru\"}, 
29
\"James Egonyu\"}, {\"collaborator\": \"John Olukuru\"}, 
            
30
{\"collaborator\": \"Chrysantus M. Tanga\"}]",
30
{\"collaborator\": \"Chrysantus M. Tanga\"}]",
            
31
  "contact_person": "Kennedy Senagi",
31
  "contact_person": "Kennedy Senagi",
            
32
  "contact_person_email": "ksenagi@icipe.org",
32
  "contact_person_email": "ksenagi@icipe.org",
            
33
  "country": "[{\"country\": \"KE\"}]",
33
  "country": "[{\"country\": \"KE\"}]",
            
34
  "creator_user_id": "f09ec764-fe3c-4069-850b-f968ff0c20bb",
34
  "creator_user_id": "f09ec764-fe3c-4069-850b-f968ff0c20bb",
            
35
  "date_uploaded": "2024-04-26",
35
  "date_uploaded": "2024-04-26",
            
36
  "donor": "Kenya Education Network Trust (KENET)",
36
  "donor": "Kenya Education Network Trust (KENET)",
            
37
  "end_date": "2024-04-26",
37
  "end_date": "2024-04-26",
            
38
  "groups": [],
38
  "groups": [],
            
39
  "id": "59311393-3b0b-4f79-92b3-62a9b30a89ff",
39
  "id": "59311393-3b0b-4f79-92b3-62a9b30a89ff",
            
40
  "isopen": false,
40
  "isopen": false,
            
41
  "license_id": "cc-nc",
41
  "license_id": "cc-nc",
            
42
  "license_title": "Creative Commons Non-Commercial (Any)",
42
  "license_title": "Creative Commons Non-Commercial (Any)",
            
43
  "license_url": "http://creativecommons.org/licenses/by-nc/2.0/",
43
  "license_url": "http://creativecommons.org/licenses/by-nc/2.0/",
            
44
  "maintainer": "Kennedy Senagi",
44
  "maintainer": "Kennedy Senagi",
            
45
  "maintainer_email": "ksenagi@icipe.org",
45
  "maintainer_email": "ksenagi@icipe.org",
            
46
  "metadata_created": "2024-04-29T09:38:54.556778",
46
  "metadata_created": "2024-04-29T09:38:54.556778",
            
n
47
  "metadata_modified": "2024-04-29T09:41:11.270664",
n
47
  "metadata_modified": "2024-04-29T09:42:08.730244",
            
48
  "name": 
48
  "name": 
            
49
p-learning-algorithms-in-farming-edible-crickets-as-a-source-of-food",
49
p-learning-algorithms-in-farming-edible-crickets-as-a-source-of-food",
            
50
  "notes": "Crickets (Gryllus bimaculatus) produce sounds as a natural 
50
  "notes": "Crickets (Gryllus bimaculatus) produce sounds as a natural 
            
51
means to communicate and convey various behaviors and activities, 
51
means to communicate and convey various behaviors and activities, 
            
52
including mating, feeding, aggression, distress, and more. These 
52
including mating, feeding, aggression, distress, and more. These 
            
53
vocalizations are intricately linked to prevailing environmental 
53
vocalizations are intricately linked to prevailing environmental 
            
54
conditions such as temperature and humidity. By accurately monitoring, 
54
conditions such as temperature and humidity. By accurately monitoring, 
            
55
identifying, and appropriately addressing these behaviors and 
55
identifying, and appropriately addressing these behaviors and 
            
56
activities, the farming and production of crickets can be enhanced. 
56
activities, the farming and production of crickets can be enhanced. 
            
57
This research implemented a decision support system that leverages 
57
This research implemented a decision support system that leverages 
            
58
machine learning (ML) algorithms to decode and classify cricket songs, 
58
machine learning (ML) algorithms to decode and classify cricket songs, 
            
59
along with their associated key weather variables (temperature and 
59
along with their associated key weather variables (temperature and 
            
60
humidity). Videos capturing cricket behavior and weather variables 
60
humidity). Videos capturing cricket behavior and weather variables 
            
61
were recorded. From these videos, sound signals were extracted and 
61
were recorded. From these videos, sound signals were extracted and 
            
62
classified such as calling, aggression, and courtship. Numerical and 
62
classified such as calling, aggression, and courtship. Numerical and 
            
63
image features were extracted from the sound signals and combined with 
63
image features were extracted from the sound signals and combined with 
            
64
the weather variables. The extracted numerical features, i.e., 
64
the weather variables. The extracted numerical features, i.e., 
            
65
Mel-Frequency Cepstral Coefficients (MFCC), Linear Frequency Cepstral 
65
Mel-Frequency Cepstral Coefficients (MFCC), Linear Frequency Cepstral 
            
66
Coefficients, and chroma, were used to train shallow (support vector 
66
Coefficients, and chroma, were used to train shallow (support vector 
            
67
machine, k-nearest neighbors, and random forest (RF)) ML algorithms. 
67
machine, k-nearest neighbors, and random forest (RF)) ML algorithms. 
            
68
While image features, i.e., spectrograms, were used to train different 
68
While image features, i.e., spectrograms, were used to train different 
            
69
state-of-the-art deep ML models, i,e., convolutional neural network} 
69
state-of-the-art deep ML models, i,e., convolutional neural network} 
            
70
architectures (ResNet152V2, VGG16, and EfficientNetB4). In the deep ML 
70
architectures (ResNet152V2, VGG16, and EfficientNetB4). In the deep ML 
            
71
category, ResNet152V2 had the best accuracy of 99.42%. The RF 
71
category, ResNet152V2 had the best accuracy of 99.42%. The RF 
            
72
algorithm had the best accuracy of 95.63% in the shallow ML category 
72
algorithm had the best accuracy of 95.63% in the shallow ML category 
            
73
when trained with a combination of MFCC+chroma and after feature 
73
when trained with a combination of MFCC+chroma and after feature 
            
74
selection. In descending order of importance, the top 6 ranked 
74
selection. In descending order of importance, the top 6 ranked 
            
75
features in the RF algorithm were, namely humidity, temperature, C#, 
75
features in the RF algorithm were, namely humidity, temperature, C#, 
            
76
mfcc11, mfcc10, and D. From the selected features. With this 
76
mfcc11, mfcc10, and D. From the selected features. With this 
            
77
information, it is notable that insects require specific temperatures 
77
information, it is notable that insects require specific temperatures 
            
78
and humidity for growth and metabolic activities. Moreover, the songs 
78
and humidity for growth and metabolic activities. Moreover, the songs 
            
79
produced by certain cricket species naturally align to musical tones 
79
produced by certain cricket species naturally align to musical tones 
            
80
such as C# and D as ranked by the algorithm. Using this knowledge, a 
80
such as C# and D as ranked by the algorithm. Using this knowledge, a 
            
81
decision support system was built to guide farmers about the optimal 
81
decision support system was built to guide farmers about the optimal 
            
82
temperature and humidity ranges and interpret the songs (calling, 
82
temperature and humidity ranges and interpret the songs (calling, 
            
83
aggression, and courtship) in relation to weather variables. With this 
83
aggression, and courtship) in relation to weather variables. With this 
            
84
information, farmers can put in place suitable measures such as 
84
information, farmers can put in place suitable measures such as 
            
85
temperature regulation, humidity control, addressing aggressors, and 
85
temperature regulation, humidity control, addressing aggressors, and 
            
86
other relevant interventions to minimize or eliminate losses and 
86
other relevant interventions to minimize or eliminate losses and 
            
87
enhance cricket production.",
87
enhance cricket production.",
            
n
88
  "num_resources": 2,
n
88
  "num_resources": 3,
            
89
  "num_tags": 6,
89
  "num_tags": 6,
            
90
  "organization": {
90
  "organization": {
            
91
    "approval_status": "approved",
91
    "approval_status": "approved",
            
92
    "created": "2022-05-17T13:53:24.098004",
92
    "created": "2022-05-17T13:53:24.098004",
            
93
    "description": "",
93
    "description": "",
            
94
    "id": "d16ea320-c49c-4a2e-b419-49c90c384c7d",
94
    "id": "d16ea320-c49c-4a2e-b419-49c90c384c7d",
            
95
    "image_url": "2022-05-17-135324.083479bees.jpeg",
95
    "image_url": "2022-05-17-135324.083479bees.jpeg",
            
96
    "is_organization": true,
96
    "is_organization": true,
            
97
    "name": "environmental-health",
97
    "name": "environmental-health",
            
98
    "state": "active",
98
    "state": "active",
            
99
    "title": "Environmental Health",
99
    "title": "Environmental Health",
            
100
    "type": "organization"
100
    "type": "organization"
            
101
  },
101
  },
            
102
  "owner_org": "d16ea320-c49c-4a2e-b419-49c90c384c7d",
102
  "owner_org": "d16ea320-c49c-4a2e-b419-49c90c384c7d",
            
103
  "principal_investigator": "Kennedy Senagi",
103
  "principal_investigator": "Kennedy Senagi",
            
104
  "principal_investigator_email": "ksenagi@icipe.org",
104
  "principal_investigator_email": "ksenagi@icipe.org",
            
105
  "private": false,
105
  "private": false,
            
106
  "region": "Nairobi",
106
  "region": "Nairobi",
            
107
  "relationships_as_object": [],
107
  "relationships_as_object": [],
            
108
  "relationships_as_subject": [],
108
  "relationships_as_subject": [],
            
109
  "resources": [
109
  "resources": [
            
110
    {
110
    {
            
111
      "cache_last_updated": null,
111
      "cache_last_updated": null,
            
112
      "cache_url": null,
112
      "cache_url": null,
            
113
      "created": "2024-04-29T09:40:16.621911",
113
      "created": "2024-04-29T09:40:16.621911",
            
114
      "date_last_updated": "2024-04-26",
114
      "date_last_updated": "2024-04-26",
            
115
      "date_uploaded_resource": "2024-04-26",
115
      "date_uploaded_resource": "2024-04-26",
            
116
      "description": "",
116
      "description": "",
            
117
      "format": "CSV",
117
      "format": "CSV",
            
118
      "hash": "",
118
      "hash": "",
            
119
      "id": "91644e4e-7951-4ed4-8533-c52c7451e557",
119
      "id": "91644e4e-7951-4ed4-8533-c52c7451e557",
            
120
      "last_modified": "2024-04-29T09:40:16.569668",
120
      "last_modified": "2024-04-29T09:40:16.569668",
            
121
      "metadata_modified": "2024-04-29T09:40:16.591494",
121
      "metadata_modified": "2024-04-29T09:40:16.591494",
            
122
      "mimetype": "text/csv",
122
      "mimetype": "text/csv",
            
123
      "mimetype_inner": null,
123
      "mimetype_inner": null,
            
124
      "name": "Main metadata sheet",
124
      "name": "Main metadata sheet",
            
125
      "package_id": "59311393-3b0b-4f79-92b3-62a9b30a89ff",
125
      "package_id": "59311393-3b0b-4f79-92b3-62a9b30a89ff",
            
126
      "position": 0,
126
      "position": 0,
            
127
      "resource_type": null,
127
      "resource_type": null,
            
128
      "restricted": "{\"allowed_users\": \"\", \"level\": 
128
      "restricted": "{\"allowed_users\": \"\", \"level\": 
            
129
\"public\"}",
129
\"public\"}",
            
130
      "size": 58048,
130
      "size": 58048,
            
131
      "state": "active",
131
      "state": "active",
            
132
      "url": 
132
      "url": 
            
133
1644e4e-7951-4ed4-8533-c52c7451e557/download/main-metadata-sheet.csv",
133
1644e4e-7951-4ed4-8533-c52c7451e557/download/main-metadata-sheet.csv",
            
134
      "url_type": "upload"
134
      "url_type": "upload"
            
135
    },
135
    },
            
136
    {
136
    {
            
137
      "cache_last_updated": null,
137
      "cache_last_updated": null,
            
138
      "cache_url": null,
138
      "cache_url": null,
            
139
      "created": "2024-04-29T09:41:11.304003",
139
      "created": "2024-04-29T09:41:11.304003",
            
140
      "date_last_updated": "2024-04-26",
140
      "date_last_updated": "2024-04-26",
            
141
      "date_uploaded_resource": "2024-04-26",
141
      "date_uploaded_resource": "2024-04-26",
            
142
      "description": "",
142
      "description": "",
            
143
      "format": "CSV",
143
      "format": "CSV",
            
144
      "hash": "",
144
      "hash": "",
            
145
      "id": "a6ea37de-2203-400f-8bd1-49e3a7d84923",
145
      "id": "a6ea37de-2203-400f-8bd1-49e3a7d84923",
            
146
      "last_modified": "2024-04-29T09:41:11.256196",
146
      "last_modified": "2024-04-29T09:41:11.256196",
            
147
      "metadata_modified": "2024-04-29T09:41:11.276536",
147
      "metadata_modified": "2024-04-29T09:41:11.276536",
            
148
      "mimetype": "text/csv",
148
      "mimetype": "text/csv",
            
149
      "mimetype_inner": null,
149
      "mimetype_inner": null,
            
150
      "name": "Chroma + weather features, and lables",
150
      "name": "Chroma + weather features, and lables",
            
151
      "package_id": "59311393-3b0b-4f79-92b3-62a9b30a89ff",
151
      "package_id": "59311393-3b0b-4f79-92b3-62a9b30a89ff",
            
152
      "position": 1,
152
      "position": 1,
            
153
      "resource_type": null,
153
      "resource_type": null,
            
154
      "restricted": "{\"allowed_users\": \"\", \"level\": 
154
      "restricted": "{\"allowed_users\": \"\", \"level\": 
            
155
\"public\"}",
155
\"public\"}",
            
156
      "size": 2244601,
156
      "size": 2244601,
            
157
      "state": "active",
157
      "state": "active",
            
158
      "url": 
158
      "url": 
            
159
0f-8bd1-49e3a7d84923/download/chroma-weather-features-and-lables.csv",
159
0f-8bd1-49e3a7d84923/download/chroma-weather-features-and-lables.csv",
            
160
      "url_type": "upload"
160
      "url_type": "upload"
            
t
t
161
    },
            
162
    {
            
163
      "cache_last_updated": null,
            
164
      "cache_url": null,
            
165
      "created": "2024-04-29T09:42:08.763024",
            
166
      "date_last_updated": "2024-04-26",
            
167
      "date_uploaded_resource": "2024-04-26",
            
168
      "description": "",
            
169
      "format": "CSV",
            
170
      "hash": "",
            
171
      "id": "81a9f5ba-c7e4-4ece-ac39-8a4f2c10089b",
            
172
      "last_modified": "2024-04-29T09:42:08.712541",
            
173
      "metadata_modified": "2024-04-29T09:42:08.736301",
            
174
      "mimetype": "text/csv",
            
175
      "mimetype_inner": null,
            
176
      "name": "MFCC + weather features, and lables",
            
177
      "package_id": "59311393-3b0b-4f79-92b3-62a9b30a89ff",
            
178
      "position": 2,
            
179
      "resource_type": null,
            
180
      "restricted": "{\"allowed_users\": \"\", \"level\": 
            
181
\"public\"}",
            
182
      "size": 1748935,
            
183
      "state": "active",
            
184
      "url": 
            
185
4ece-ac39-8a4f2c10089b/download/mfcc-weather-features-and-lables.csv",
            
186
      "url_type": "upload"
            
161
    }
187
    }
            
162
  ],
188
  ],
            
163
  "start_date": "2024-04-26",
189
  "start_date": "2024-04-26",
            
164
  "state": "draft",
190
  "state": "draft",
            
165
  "tags": [
191
  "tags": [
            
166
    {
192
    {
            
167
      "display_name": "Insects",
193
      "display_name": "Insects",
            
168
      "id": "9c6eae2d-7311-41e5-958f-3b11a4a905d8",
194
      "id": "9c6eae2d-7311-41e5-958f-3b11a4a905d8",
            
169
      "name": "Insects",
195
      "name": "Insects",
            
170
      "state": "active",
196
      "state": "active",
            
171
      "vocabulary_id": null
197
      "vocabulary_id": null
            
172
    },
198
    },
            
173
    {
199
    {
            
174
      "display_name": "decision support system",
200
      "display_name": "decision support system",
            
175
      "id": "22b7773d-18d8-4b5d-a86d-7058197312a0",
201
      "id": "22b7773d-18d8-4b5d-a86d-7058197312a0",
            
176
      "name": "decision support system",
202
      "name": "decision support system",
            
177
      "state": "active",
203
      "state": "active",
            
178
      "vocabulary_id": null
204
      "vocabulary_id": null
            
179
    },
205
    },
            
180
    {
206
    {
            
181
      "display_name": "deep learning",
207
      "display_name": "deep learning",
            
182
      "id": "0265fb58-9cd3-4d0d-8215-2f0b571a09f4",
208
      "id": "0265fb58-9cd3-4d0d-8215-2f0b571a09f4",
            
183
      "name": "deep learning",
209
      "name": "deep learning",
            
184
      "state": "active",
210
      "state": "active",
            
185
      "vocabulary_id": null
211
      "vocabulary_id": null
            
186
    },
212
    },
            
187
    {
213
    {
            
188
      "display_name": "machine learning",
214
      "display_name": "machine learning",
            
189
      "id": "3638359c-f3cb-452c-ac1a-57193f56c880",
215
      "id": "3638359c-f3cb-452c-ac1a-57193f56c880",
            
190
      "name": "machine learning",
216
      "name": "machine learning",
            
191
      "state": "active",
217
      "state": "active",
            
192
      "vocabulary_id": null
218
      "vocabulary_id": null
            
193
    },
219
    },
            
194
    {
220
    {
            
195
      "display_name": "sound classification",
221
      "display_name": "sound classification",
            
196
      "id": "e5acbf75-416c-4213-abc9-bfd0a0cca7d9",
222
      "id": "e5acbf75-416c-4213-abc9-bfd0a0cca7d9",
            
197
      "name": "sound classification",
223
      "name": "sound classification",
            
198
      "state": "active",
224
      "state": "active",
            
199
      "vocabulary_id": null
225
      "vocabulary_id": null
            
200
    },
226
    },
            
201
    {
227
    {
            
202
      "display_name": "transfer learning",
228
      "display_name": "transfer learning",
            
203
      "id": "837a5abb-e47d-4c1e-88e1-5c75879de94d",
229
      "id": "837a5abb-e47d-4c1e-88e1-5c75879de94d",
            
204
      "name": "transfer learning",
230
      "name": "transfer learning",
            
205
      "state": "active",
231
      "state": "active",
            
206
      "vocabulary_id": null
232
      "vocabulary_id": null
            
207
    }
233
    }
            
208
  ],
234
  ],
            
209
  "third_party": "no",
235
  "third_party": "no",
            
210
  "title": "A Convolutional Neural Network with Image and Numerical 
236
  "title": "A Convolutional Neural Network with Image and Numerical 
            
211
Data to Improve Farming of Edible Crickets as a Source of Food - A 
237
Data to Improve Farming of Edible Crickets as a Source of Food - A 
            
212
Decision Support System",
238
Decision Support System",
            
213
  "type": "dataset",
239
  "type": "dataset",
            
214
  "url": null,
240
  "url": null,
            
215
  "version": null
241
  "version": null
            
216
}
242
}