IEEE Computational Intelligence Magazine - February 2020 - 81

of lexicons such as affective words.
T h u m b s U p -Down (TH) model,
m 3 : text " R 2, counts the number of
affective words keeping a separate record
for the positive and negative words. We
created a positive-negative lexicon based
on several affective lexicons for English
[46] and Spanish [47]-[49] and enriched
with WordNet [50]. In the case of Arabic, we translate the English lexicon to
Arabic language using Google translate,
service employing Googletrans API [51].
C. Emoji Space

Inspired by DeepMoji [15], we create
the text model, m 4 : text " R 64; the
core idea is to predict what emoji would
be the most probable one for a given
text. For this purpose, we learn a B4MSA
model per language using 3.2  million
examples of the 64 most frequent emojis
in each language. This dataset consists in
50,000 examples per emoji extracted
from our own collected tweets, that is,
we filtered out these examples from
(approximately) 2.0 # 10 9 Arabic tweets,
2.3 # 10 9 English tweets, and to
3.7 # 10 9 Spanish tweets. A few simple
rules were followed to create the datasets: i) each example contains only one
type of emoji to reduce the ambiguity
among predictions, ii) all re-tweets were
removed, iii) a uniform sample was chosen to avoid any seasonality effect. Finally, each selected tweet is transformed
into a text and emoji pair, where the
emoji is the one in the text. All emojis

(a)

were removed from the text while training. Consequently, the dataset is a supervised learning dataset.
B4MSA uses this dataset to create the
Emoji Space. Each text is transformed to
the vector space defined by B4MSA's
text model using Table I parameters
with a one-vs-rest strategy to train the
SVM (i.e., m 4 = g b % m b where m b is
B4MSA's text model and g b is a linear
SVM). Instead of being interested in the
most probable emoji given text, we are
interested in the decision functions of all
the classifiers given a text such that each
coordinate represents an emoji. Consequently, a 64-dimension real-valued vector represents a text.
Figure 2 lists the emojis used to create our Emoji-Space for Spanish, English,
and Arabic languages; which also correspond to 64 most frequent emojis in
these languages. The emojis are ordered
row-wise being the most frequent the
emoji in the left upper corner. Notice
the significant coincidence among the
most frequent emojis in all languages.
D. FastText

FastText [52] is a tool to create text classifiers and learn a semantic vocabulary from a
given collection of documents; this vocabulary is represented with a collection of
high dimensional vectors, one per word.
FastText is robust to lexical errors supporting out-vocabulary words, and it is used to
represent a text into a vector space using
the pre-computed models (see [53]) for

(b)

Arabic, English, and Spanish. In particular,
each text is transformed into a vector
using the vector sentences flag; these are
vectors in 300 dimensions using the
default parameters (i.e., m 5 : text " R 300).
E. EvoDAG

EvoDAG7 [42], [43] is a steady-state GP
system with tournament selection (tournament size 2) specifically tailored to
tackle classification and regression problems. GP is an evolutionary algorithm
with the distinctive characteristic of
searching in a program search space, in
particular, in this contribution, GP
searches in a search space, X, of functions. That is, X is the set of functions
created by recursively composing elements from two sets: function set F, and
terminal set L. The function set is composed by operations such as sum, product, sin, cos, max, and min, among others;
and the inputs compose the terminal set,
and normally, by an ephemeral random
constant. Nonetheless, EvoDAG's terminal set only contains inputs, and each
function, in the function set, is associated
with a set of parameters that are identified using the training set. For example,
let f ! F be a function of cardinality 1
then f (x ; i) ! X is an element of the
search space, and, i is identified with the
training set using ordinary least squares,
e.g., f (x ; i) = i sin (x).

7

https://github.com/mgraffg/EvoDAG

(c)

FIGURE 2 Actual emojis used by our Emoji-Space model. (a) Spanish, (b) English, (c) Arabic.

FEBRUARY 2020 | IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE

81
https://github.com/mgraffg/EvoDAG
IEEE Computational Intelligence Magazine - February 2020

Table of Contents for the Digital Edition of IEEE Computational Intelligence Magazine - February 2020
