LONDON — As any dog owner knows, our furry friends have unique personalities that deeply impact the human-animal bond. Now, an ingenious new artificial intelligence (AI) system can analyze a dog’s behavior and predict their basic personality type with up to 99 percent accuracy.
Developed by computer scientists at the University of East London and the University of Pennsylvania, this groundbreaking technology called “C-BARQ” uses machine learning algorithms to classify dogs into one of five personality categories based on their responses to everyday situations. The researchers say this AI profiling tool could revolutionize how dogs are selected and trained for specialized working roles such as guide, assistance, detection, and search and rescue dogs. It may also improve the pet adoption process by personality-matching dogs with compatible owners.
The research was commissioned by Dogvatar, a Miami-based canine technology startup.
Cracking the Canine Personality Code
Previous studies have tried to develop standardized personality tests for dogs using owner surveys, but none have taken an AI approach until now.
The study authors used the Canine Behavioral Assessment and Research Questionnaire (C-BARQ) database from the University of Pennsylvania School of Veterinary Medicine, containing behavioral records for over 70,000 dogs. The C-BARQ survey asks owners to rate the frequency and severity of their dog’s behavioral responses to 100 common situations on a scale from 0 to 4.
“C-BARQ is highly effective, but many of its questions are also subjective,” says co-Principal Investigator James Serpell, a professor of ethics and animal welfare emeritus at the UPenn School of Veterinary Medicine, in a statement. “By clustering data from thousands of surveys, we can adjust for outlying responses inherent to subjective survey questions in categories such as dog rivalry and stranger-directed fear.”
The program automatically clustered the dogs into five distinct personality categories when the team ran an unsupervised machine-learning algorithm called K-Means on the C-BARQ data. By analyzing the strength of associations between behaviors and categories, descriptive labels were assigned to each of the five groupings:
1. Excitable/Hyperattached – Energetic, attached, playful, hard to settle down
2. Anxious/Fearful – Shy, startles easily, fearful of unfamiliar people or dogs
3. Aloof/Predatory – Low attachment/attention-seeking, high prey drive
4. Reactive/Assertive – Aggressive tendencies, reacts strongly to discipline
5. Calm/Agreeable – Low aggression/fear, eager to please, fast learner
To evaluate the AI system’s predictive power, the researchers tested how accurately four machine learning models could assign individual dogs to one of the five personality types based only on their C-BARQ responses. The best-performing model, called a Decision Tree, achieved near-perfect predictions with 99 percent accuracy. Support Vector Machine and K-Nearest Neighbor models scored 98 percent and 97 percent, respectively, while the Naïve Bayes model lagged at 77 percent accuracy.
Why is nailing down canine personality traits so crucial in the first place? Researchers say dogs with undesirable personality traits are at much higher risk of being surrendered to animal shelters. In specialized working dogs like guide dogs, certain traits make them far more successful than others at their jobs. The policy implications of this new technology are, therefore, far-reaching.
Applications: Choosing Fido for the Job
The researchers suggest AI-based canine personality profiling could revolutionize how prospective working dogs are selected and trained for roles like:
• Guide dogs – Calm/Agreeable personalities fare best
• Assistance dogs – Excitable/Hyperattached types preferred over Anxious/Fearful
• Drug/bomb detection dogs – Aloof/Predatory personalities excel
• Search and rescue dogs – Highly Driven, Reactive/Assertive types succeed
It could also improve the selection of dogs for specific sports like agility, racing, or herding competitions where drive, athleticism, and competitiveness are vital.
The technology may additionally assist animal shelters and rescue and adoption agencies to make better initial matches between available dogs and prospective owners. Compatibility between pet personality and owner lifestyle/experience level often determines whether placements succeed long-term. Now, we have an evidence-based tool to find our canine companions.
“This has been a really exciting breakthrough for us,” says Dogvatar CEO “Alpha Pack Leader” Piya Pettigrew. “This algorithm could greatly improve efficiency in the working dog training and placement process, and could help reduce the number of companion dogs brought back to shelters for not being compatible. It’s a win for both dogs and the people they serve.”
While optimistic about AI-powered canine personality profiling, there are limitations. The current C-BARQ database relies wholly on owner surveys – subjective impressions that could bias predictions. The team is now working to validate their AI models using more objective indicators of canine temperament like behavioral tests, activity monitors, physiological stress biomarkers (e.g., heart rate, cortisol levels), and genetics. Integrating these concrete metrics with owner reports will likely further enhance predictive accuracy.
As machine learning and “doggynomics” continue advancing hand-in-paw, artificially intelligent match-making for mixed-breed shelter pups may one day be as common as online dating sites for humans. Man’s best friends can look forward to a brighter future where compatibility reigns supreme!
The research is published in Scientific Reports.
You might also be interested in: