Changes

              
    

          
          
        
        

            
f
1
{
f
1
{
            
2
  "acknowledgement": "",
2
  "acknowledgement": "",
            
3
  "administrative_areas": "",
3
  "administrative_areas": "",
            
4
  "author": null,
4
  "author": null,
            
5
  "author_email": null,
5
  "author_email": null,
            
6
  "citation_narrative": "",
6
  "citation_narrative": "",
            
7
  "collaborators": "[{\"collaborator\": \"Fatiah Khamis\"}]",
7
  "collaborators": "[{\"collaborator\": \"Fatiah Khamis\"}]",
            
8
  "contact_person": "Daisy Salifu",
8
  "contact_person": "Daisy Salifu",
            
9
  "contact_person_email": "dsalifu@icipe.org",
9
  "contact_person_email": "dsalifu@icipe.org",
            
10
  "country": "[{\"country\": \"KE\"}]",
10
  "country": "[{\"country\": \"KE\"}]",
            
11
  "creator_user_id": "f09ec764-fe3c-4069-850b-f968ff0c20bb",
11
  "creator_user_id": "f09ec764-fe3c-4069-850b-f968ff0c20bb",
            
12
  "date_uploaded": "2021-10-17",
12
  "date_uploaded": "2021-10-17",
            
13
  "donor": "",
13
  "donor": "",
            
14
  "groups": [],
14
  "groups": [],
            
15
  "id": "4bd2ffd3-fc6c-4189-bce1-f54ca34ea011",
15
  "id": "4bd2ffd3-fc6c-4189-bce1-f54ca34ea011",
            
16
  "isopen": true,
16
  "isopen": true,
            
17
  "license_id": "cc-by",
17
  "license_id": "cc-by",
            
18
  "license_title": "Creative Commons Attribution",
18
  "license_title": "Creative Commons Attribution",
            
19
  "license_url": "http://www.opendefinition.org/licenses/cc-by",
19
  "license_url": "http://www.opendefinition.org/licenses/cc-by",
            
20
  "maintainer": "Kennedy Senagi",
20
  "maintainer": "Kennedy Senagi",
            
21
  "maintainer_email": "ksenagi@icipe.org",
21
  "maintainer_email": "ksenagi@icipe.org",
            
22
  "metadata_created": "2022-05-18T05:49:38.498118",
22
  "metadata_created": "2022-05-18T05:49:38.498118",
            
n
23
  "metadata_modified": "2022-05-25T12:11:02.211131",
n
23
  "metadata_modified": "2022-07-21T10:03:12.235468",
            
24
  "name": "african-fruit-fly-program",
24
  "name": "african-fruit-fly-program",
            
25
  "notes": "Taxonomic identity of fruitfly pest using morphometrics",
25
  "notes": "Taxonomic identity of fruitfly pest using morphometrics",
            
26
  "num_resources": 4,
26
  "num_resources": 4,
            
27
  "num_tags": 0,
27
  "num_tags": 0,
            
28
  "organization": {
28
  "organization": {
            
29
    "approval_status": "approved",
29
    "approval_status": "approved",
            
30
    "created": "2022-05-17T13:51:43.824504",
30
    "created": "2022-05-17T13:51:43.824504",
            
31
    "description": "",
31
    "description": "",
            
32
    "id": "b3f6f7ff-df08-49a0-8959-dc1444685c2f",
32
    "id": "b3f6f7ff-df08-49a0-8959-dc1444685c2f",
            
33
    "image_url": "2022-05-17-135143.809085plant.jpeg",
33
    "image_url": "2022-05-17-135143.809085plant.jpeg",
            
34
    "is_organization": true,
34
    "is_organization": true,
            
35
    "name": "plant-health",
35
    "name": "plant-health",
            
36
    "state": "active",
36
    "state": "active",
            
37
    "title": "Plant Health",
37
    "title": "Plant Health",
            
38
    "type": "organization"
38
    "type": "organization"
            
39
  },
39
  },
            
40
  "owner_org": "b3f6f7ff-df08-49a0-8959-dc1444685c2f",
40
  "owner_org": "b3f6f7ff-df08-49a0-8959-dc1444685c2f",
            
41
  "principal_investigator": "Daniel Masiga",
41
  "principal_investigator": "Daniel Masiga",
            
42
  "principal_investigator_email": "dmasiga@icipe.org",
42
  "principal_investigator_email": "dmasiga@icipe.org",
            
43
  "private": false,
43
  "private": false,
            
44
  "region": "East Africa",
44
  "region": "East Africa",
            
45
  "relationships_as_object": [],
45
  "relationships_as_object": [],
            
46
  "relationships_as_subject": [],
46
  "relationships_as_subject": [],
            
47
  "resources": [
47
  "resources": [
            
48
    {
48
    {
            
49
      "cache_last_updated": null,
49
      "cache_last_updated": null,
            
50
      "cache_url": null,
50
      "cache_url": null,
            
51
      "created": "2022-05-18T05:51:33.159984",
51
      "created": "2022-05-18T05:51:33.159984",
            
52
      "date_last_updated": "2022-05-18",
52
      "date_last_updated": "2022-05-18",
            
53
      "date_uploaded_resource": "2021-10-17",
53
      "date_uploaded_resource": "2021-10-17",
            
54
      "description": "",
54
      "description": "",
            
55
      "format": "XLSX",
55
      "format": "XLSX",
            
56
      "hash": "",
56
      "hash": "",
            
57
      "id": "0613d1d8-d0a2-4d34-9b4d-29f698855766",
57
      "id": "0613d1d8-d0a2-4d34-9b4d-29f698855766",
            
58
      "last_modified": "2022-05-18T05:51:33.141075",
58
      "last_modified": "2022-05-18T05:51:33.141075",
            
59
      "metadata_modified": "2022-05-23T08:23:57.289435",
59
      "metadata_modified": "2022-05-23T08:23:57.289435",
            
60
      "mimetype": 
60
      "mimetype": 
            
61
"application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
61
"application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
            
62
      "mimetype_inner": null,
62
      "mimetype_inner": null,
            
63
      "name": "Raw data, metadata and data dictionary ",
63
      "name": "Raw data, metadata and data dictionary ",
            
64
      "package_id": "4bd2ffd3-fc6c-4189-bce1-f54ca34ea011",
64
      "package_id": "4bd2ffd3-fc6c-4189-bce1-f54ca34ea011",
            
65
      "position": 0,
65
      "position": 0,
            
66
      "resource_type": null,
66
      "resource_type": null,
            
67
      "restricted": "{\"allowed_users\": \"\", \"level\": 
67
      "restricted": "{\"allowed_users\": \"\", \"level\": 
            
68
\"public\"}",
68
\"public\"}",
            
69
      "size": 734510,
69
      "size": 734510,
            
70
      "state": "active",
70
      "state": "active",
            
71
      "url": 
71
      "url": 
            
72
-29f698855766/download/fruitfly-morphometrics_data-repository-1.xlsx",
72
-29f698855766/download/fruitfly-morphometrics_data-repository-1.xlsx",
            
73
      "url_type": "upload"
73
      "url_type": "upload"
            
74
    },
74
    },
            
75
    {
75
    {
            
76
      "cache_last_updated": null,
76
      "cache_last_updated": null,
            
77
      "cache_url": null,
77
      "cache_url": null,
            
78
      "created": "2022-05-18T05:59:22.824877",
78
      "created": "2022-05-18T05:59:22.824877",
            
79
      "date_last_updated": "2022-05-18",
79
      "date_last_updated": "2022-05-18",
            
80
      "date_uploaded_resource": "2021-10-17",
80
      "date_uploaded_resource": "2021-10-17",
            
81
      "description": "",
81
      "description": "",
            
82
      "format": "PDF",
82
      "format": "PDF",
            
83
      "hash": "",
83
      "hash": "",
            
84
      "id": "7524a0ed-79fc-47cf-a478-fb0b45988fe1",
84
      "id": "7524a0ed-79fc-47cf-a478-fb0b45988fe1",
            
85
      "last_modified": "2022-05-18T05:59:22.800928",
85
      "last_modified": "2022-05-18T05:59:22.800928",
            
86
      "metadata_modified": "2022-05-18T05:59:22.819494",
86
      "metadata_modified": "2022-05-18T05:59:22.819494",
            
87
      "mimetype": "application/pdf",
87
      "mimetype": "application/pdf",
            
88
      "mimetype_inner": null,
88
      "mimetype_inner": null,
            
89
      "name": "Protocol",
89
      "name": "Protocol",
            
90
      "package_id": "4bd2ffd3-fc6c-4189-bce1-f54ca34ea011",
90
      "package_id": "4bd2ffd3-fc6c-4189-bce1-f54ca34ea011",
            
91
      "position": 1,
91
      "position": 1,
            
92
      "resource_type": null,
92
      "resource_type": null,
            
93
      "restricted": "{\"allowed_users\": \"\", \"level\": 
93
      "restricted": "{\"allowed_users\": \"\", \"level\": 
            
94
\"public\"}",
94
\"public\"}",
            
95
      "size": 456159,
95
      "size": 456159,
            
96
      "state": "active",
96
      "state": "active",
            
97
      "url": 
97
      "url": 
            
98
/resource/7524a0ed-79fc-47cf-a478-fb0b45988fe1/download/protocol.pdf",
98
/resource/7524a0ed-79fc-47cf-a478-fb0b45988fe1/download/protocol.pdf",
            
99
      "url_type": "upload"
99
      "url_type": "upload"
            
100
    },
100
    },
            
101
    {
101
    {
            
102
      "cache_last_updated": null,
102
      "cache_last_updated": null,
            
103
      "cache_url": null,
103
      "cache_url": null,
            
104
      "created": "2022-05-18T05:54:07.020692",
104
      "created": "2022-05-18T05:54:07.020692",
            
105
      "date_last_updated": "2022-05-18",
105
      "date_last_updated": "2022-05-18",
            
106
      "date_uploaded_resource": "2021-10-17",
106
      "date_uploaded_resource": "2021-10-17",
            
107
      "description": "",
107
      "description": "",
            
108
      "format": "HTML",
108
      "format": "HTML",
            
109
      "hash": "",
109
      "hash": "",
            
110
      "id": "e8e406c3-6816-4330-99b4-8ef68999eb76",
110
      "id": "e8e406c3-6816-4330-99b4-8ef68999eb76",
            
111
      "last_modified": null,
111
      "last_modified": null,
            
112
      "metadata_modified": "2022-05-18T05:54:07.015551",
112
      "metadata_modified": "2022-05-18T05:54:07.015551",
            
113
      "mimetype": null,
113
      "mimetype": null,
            
114
      "mimetype_inner": null,
114
      "mimetype_inner": null,
            
115
      "name": "[R codes] - Machine learning algorithms on insect 
115
      "name": "[R codes] - Machine learning algorithms on insect 
            
116
morphometrics data",
116
morphometrics data",
            
117
      "package_id": "4bd2ffd3-fc6c-4189-bce1-f54ca34ea011",
117
      "package_id": "4bd2ffd3-fc6c-4189-bce1-f54ca34ea011",
            
118
      "position": 2,
118
      "position": 2,
            
119
      "resource_type": null,
119
      "resource_type": null,
            
120
      "restricted": "{\"allowed_users\": \"\", \"level\": 
120
      "restricted": "{\"allowed_users\": \"\", \"level\": 
            
121
\"public\"}",
121
\"public\"}",
            
122
      "size": null,
122
      "size": null,
            
123
      "state": "active",
123
      "state": "active",
            
124
      "url": 
124
      "url": 
            
125
pe-official/Machine-Learning-Algorithms-on-Insect-Morphometrics-Data",
125
pe-official/Machine-Learning-Algorithms-on-Insect-Morphometrics-Data",
            
126
      "url_type": null
126
      "url_type": null
            
127
    },
127
    },
            
128
    {
128
    {
            
129
      "cache_last_updated": null,
129
      "cache_last_updated": null,
            
130
      "cache_url": null,
130
      "cache_url": null,
            
131
      "created": "2022-05-18T05:55:37.808750",
131
      "created": "2022-05-18T05:55:37.808750",
            
132
      "date_last_updated": "2022-05-18",
132
      "date_last_updated": "2022-05-18",
            
133
      "date_uploaded_resource": "2022-05-18",
133
      "date_uploaded_resource": "2022-05-18",
            
134
      "description": "Analysis of landmark-based morphometric 
134
      "description": "Analysis of landmark-based morphometric 
            
135
measurements taken on body parts of insects have been a useful 
135
measurements taken on body parts of insects have been a useful 
            
136
taxonomic approach alongside DNA barcoding in insect identification. 
136
taxonomic approach alongside DNA barcoding in insect identification. 
            
137
Statistical analysis of morphometrics have largely been dominated by 
137
Statistical analysis of morphometrics have largely been dominated by 
            
138
traditional methods and approaches such as principal component 
138
traditional methods and approaches such as principal component 
            
139
analysis (PCA), canonical variate analysis (CVA) and discriminant 
139
analysis (PCA), canonical variate analysis (CVA) and discriminant 
            
140
analysis (DA). However, advancement in computing power creates a 
140
analysis (DA). However, advancement in computing power creates a 
            
141
paradigm shift to apply modern tools such as machine learning. Herein, 
141
paradigm shift to apply modern tools such as machine learning. Herein, 
            
142
we assess the predictive performance of four machine learning 
142
we assess the predictive performance of four machine learning 
            
143
classifiers; K-nearest neighbor (KNN), random forest (RF), support 
143
classifiers; K-nearest neighbor (KNN), random forest (RF), support 
            
144
vector machine (the linear, polynomial and radial kernel SVMs) and 
144
vector machine (the linear, polynomial and radial kernel SVMs) and 
            
145
artificial neural network (ANNs) on fruit fly morphometrics that were 
145
artificial neural network (ANNs) on fruit fly morphometrics that were 
            
146
previously analysed using PCA and CVA. KNN and RF performed poorly 
146
previously analysed using PCA and CVA. KNN and RF performed poorly 
            
147
with overall model accuracy lower than \u201cno-information rate\u201d 
147
with overall model accuracy lower than \u201cno-information rate\u201d 
            
148
(NIR) (p value\u2009>\u20090.1). The SVM models had a predictive 
148
(NIR) (p value\u2009>\u20090.1). The SVM models had a predictive 
            
149
accuracy of\u2009>\u200995%, significantly higher than NIR 
149
accuracy of\u2009>\u200995%, significantly higher than NIR 
            
150
(p\u2009<\u20090.001), Kappa\u2009>\u20090.78 and area under curve 
150
(p\u2009<\u20090.001), Kappa\u2009>\u20090.78 and area under curve 
            
151
(AUC) of the receiver operating characteristics was\u2009>\u20090.91; 
151
(AUC) of the receiver operating characteristics was\u2009>\u20090.91; 
            
152
while ANN model had a predictive accuracy of 96%, significantly higher 
152
while ANN model had a predictive accuracy of 96%, significantly higher 
            
153
than NIR, Kappa of 0.83 and AUC was 0.98. Wing veins 2, 3, 8, 10, 14 
153
than NIR, Kappa of 0.83 and AUC was 0.98. Wing veins 2, 3, 8, 10, 14 
            
154
and tibia length were of higher importance than other variables based 
154
and tibia length were of higher importance than other variables based 
            
155
on both SVM and ANN models. We conclude that SVM and ANN models could 
155
on both SVM and ANN models. We conclude that SVM and ANN models could 
            
156
be used to discriminate fruit fly species based on wing vein and tibia 
156
be used to discriminate fruit fly species based on wing vein and tibia 
            
157
length measurements or any other morphologically similar pest taxa. 
157
length measurements or any other morphologically similar pest taxa. 
            
158
These algorithms could be used as candidates for developing an 
158
These algorithms could be used as candidates for developing an 
            
159
integrated and smart application software for insect discrimination 
159
integrated and smart application software for insect discrimination 
            
160
and identification. Variable importance analysis results in this study 
160
and identification. Variable importance analysis results in this study 
            
161
would be useful for future studies for deciding what must be 
161
would be useful for future studies for deciding what must be 
            
162
measured.",
162
measured.",
            
163
      "format": "HTML",
163
      "format": "HTML",
            
164
      "hash": "",
164
      "hash": "",
            
165
      "id": "cb7e4445-4440-4098-8b72-441a2597bb47",
165
      "id": "cb7e4445-4440-4098-8b72-441a2597bb47",
            
166
      "last_modified": null,
166
      "last_modified": null,
            
n
167
      "metadata_modified": "2022-05-18T05:55:37.803541",
n
167
      "metadata_modified": "2022-07-21T10:03:12.241679",
            
168
      "mimetype": null,
168
      "mimetype": null,
            
169
      "mimetype_inner": null,
169
      "mimetype_inner": null,
            
170
      "name": "[Article] - Leveraging machine learning tools and 
170
      "name": "[Article] - Leveraging machine learning tools and 
            
171
algorithms for analysis of fruit fly morphometrics",
171
algorithms for analysis of fruit fly morphometrics",
            
172
      "package_id": "4bd2ffd3-fc6c-4189-bce1-f54ca34ea011",
172
      "package_id": "4bd2ffd3-fc6c-4189-bce1-f54ca34ea011",
            
173
      "position": 3,
173
      "position": 3,
            
174
      "resource_type": null,
174
      "resource_type": null,
            
175
      "restricted": "{\"allowed_users\": \"\", \"level\": 
175
      "restricted": "{\"allowed_users\": \"\", \"level\": 
            
176
\"public\"}",
176
\"public\"}",
            
177
      "size": null,
177
      "size": null,
            
178
      "state": "active",
178
      "state": "active",
            
t
179
      "url": "",
t
179
      "url": "https://www.nature.com/articles/s41598-022-11258-w",
            
180
      "url_type": null
180
      "url_type": null
            
181
    }
181
    }
            
182
  ],
182
  ],
            
183
  "state": "active",
183
  "state": "active",
            
184
  "tags": [],
184
  "tags": [],
            
185
  "third_party": "no",
185
  "third_party": "no",
            
186
  "title": "African fruit fly program",
186
  "title": "African fruit fly program",
            
187
  "type": "dataset",
187
  "type": "dataset",
            
188
  "url": null,
188
  "url": null,
            
189
  "version": null
189
  "version": null
            
190
}
190
}