Back

species2vec: A novel method for species representation

Angelov, B.

2019-12-22 ecology
10.1101/461996 bioRxiv
Show abstract

Word embeddings are omnipresent in Natural Language Processing (NLP) tasks. The same technology which defines words by their context can also define biological species. This study showcases this new method - species embedding (species2vec). By proximity sorting of 6761594 mammal observations from the whole world (2862 different species), we are able to create a training corpus for the skip-gram model. The resulting species embeddings are tested in an environmental classification task. The classifier performance confirms the utility of those embeddings in preserving the relationships between species, and also being representative of species consortia in an environment.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

1
Ecological Informatics
○ 33 papers in training set
Top 0.1%
30.8%
2
PLOS ONE
● 5266 papers in training set
Top 15%
12.3%
3
Scientific Reports
○ 3612 papers in training set
Top 21%
4.8%
4
Bioinformatics Advances
◐ 203 papers in training set
Top 1%
4.8%
50% of probability mass above
5
Ecology and Evolution
○ 267 papers in training set
Top 2%
4.0%
6
GigaScience
◐ 212 papers in training set
Top 1%
3.5%
7
PeerJ
◐ 308 papers in training set
Top 2%
3.2%
8
Methods in Ecology and Evolution
○ 176 papers in training set
Top 0.7%
2.8%
9
PLOS Computational Biology
● 1863 papers in training set
Top 11%
2.6%
10
Ecography
○ 54 papers in training set
Top 0.6%
2.4%
11
Frontiers in Ecology and Evolution
○ 69 papers in training set
Top 0.8%
2.4%
12
Ecological Indicators
○ 21 papers in training set
Top 0.3%
1.7%
13
Environmental DNA
○ 56 papers in training set
Top 0.5%
1.1%
14
Peer Community Journal
● 281 papers in training set
Top 4%
1.1%
15
Scientific Data
○ 209 papers in training set
Top 2%
1.0%
16
Royal Society Open Science
● 214 papers in training set
Top 6%
0.8%
17
Parasites & Vectors
○ 60 papers in training set
Top 1%
0.8%
18
Computational and Structural Biotechnology Journal
● 242 papers in training set
Top 7%
0.8%
19
Data in Brief
○ 14 papers in training set
Top 0.2%
0.6%
20
Viruses
○ 332 papers in training set
Top 5%
0.6%
21
Systematic Entomology
○ 14 papers in training set
Top 0.3%
0.6%
22
IEEE Access
● 35 papers in training set
Top 2%
0.6%
23
Journal of Animal Ecology
○ 75 papers in training set
Top 2%
0.6%
24
Briefings in Bioinformatics
◐ 354 papers in training set
Top 8%
0.6%
25
BMC Genomics
○ 406 papers in training set
Top 10%
0.6%
26
Sustainability
○ 10 papers in training set
Top 0.6%
0.6%
27
Gigabyte
● 62 papers in training set
Top 1%
0.6%