They present artificial intelligence capable of predicting the age and sex of a baby based on its temperament – InTallaght

During the first few months of a babys life, it is usually difficult to tell the exact sex or age by sight alone.

A machine learning model showed that it is possible to identify both indicators, the sex and the age of a baby, based on its temperament, thanks to artificial intelligence.

A babys propensity to show fear, smile or laugh can tell a lot about their age or gender, in the eyes of artificial intelligence.

This was demonstrated by a research team led by professionals from Washington State University and the University of Idaho, United States, through a study in which a machine learning model was used to analyze temperament data in 4,438 babies, in an attempt to classify babies by gender and age.

Working with this many babies is an almost intractable task for just one research laboratory. However, obtaining this sample was possible thanks to the collection of reliable information from different scientific sources, to collect child behavior data collected between 2006 and 2019.

The data with which we worked in this research come from a questionnaire that applies a temperament measurement scale, in which parents recorded the frequency of 191 different behaviors that their children from 3 to 12 months of age showed during one week. These behaviors can be categorized into 14 different dimensions of temperament, such as smiling, activity level, anger or frustration, and fear. Overall, the sample collects data from 2,298 boys and 2,093 girls.

Working with this background, an AI model capable of using these indicators to identify the age and sex of a baby was developed. Previously, research has already been carried out that, in an isolated way, investigates the relationship between temperament and the age or sex of the baby. This research, on the other hand, is the first to combine these elements during the analysis.

During the first 48 weeks of a babys life, it was easier for algorithms to decipher a babys age than its gender. However, after that period, the system was able to specify their classification, which suggests that gender differences are accentuated from that age.

It is at least suggestive of a picture in which temperament begins to differentiate by gender in a more powerful way around the age of one year, commented Maria Gartstein, lead author of the study and professor of psychology at Washington State University.

For purposes of this study, research co-author Erich Seamon of the University of Idaho Institute for Modeling, Collaboration and Innovation used machine learning algorithms to classify babies as male or female under three age ranges: 0- 24 weeks of age, 24-48 weeks of age, and older than 48 weeks, based on their scores for the 14 dimensions of temperament.

Accuracy rates increased with age, from a low of 38% for age group one to 57% for age group three.

It was a good opportunity to do a kind of demonstration study using these machine learning techniques that require really large data sets and are not very common in socio-emotional development researchGartstein commented. It gave us the opportunity for the first time to really consider the extent to which gender differences are informed by childhood age.

Although the results of this research are subject to the multiple influences that may affect its study sample, the work presented is consistent with previous research that already indicated that the effects of socialization begin to become visible around one year of age.

Read more here:

They present artificial intelligence capable of predicting the age and sex of a baby based on its temperament - InTallaght