Commercial companies have always been interested in knowing customer opinions regarding their latest advertisements on television and radio; these companies usually invite a group of potential customers (in focus groups) to gauge the success of an ad. However, this traditional method may soon be obsolete. According to New Scientist, a startup named Affectiva, located in Massachusetts, USA, claims to have developed software that can predict audience behavior regarding advertisements by reading facial movements during and immediately after the ad broadcast. Daniel McDuff, a scientist and one of the principal researchers of this project, states that companies will be able to understand what audiences are thinking at any moment during the ad broadcast, which is very different from traditional methods that ask for audience opinions after the ads are aired; 'This method provides real-time interactions from the audience. In fact, it allows us to understand what they truly feel without that feeling first going to the conscious mind for interpretation.' Affectiva's software initially identifies key facial points such as the mouth, eyebrows, and tip of the nose. It then uses specific machine learning algorithms (one of the most practical areas of artificial intelligence) to study how these points move or even changes in temperature and color of the face during the ad broadcast. These behaviors and changes in skin and facial features while watching the ad are broken down into smaller parts, allowing for the interpretation of audience emotions at those moments. In a study conducted by McDuff and his team, over 1,200 television viewers were monitored by cameras installed in their homes while watching ads related to sweets, household food items, and pet supplies. Before and after each ad broadcast, participants filled out questionnaires regarding their likelihood of purchasing the advertised items. The software measured emotions such as happiness, surprise, and confusion among viewers during the ads. After the ads concluded, the research team found that they could predict questionnaire results using their algorithm's findings and determine which viewer intended to purchase which item. The Affectiva team analyzed data from over three million videos of individuals of various ages, genders, and ethnic backgrounds. McDuff notes that 'there seem to be subtle and precise changes in emotional responses: for example, women exhibit more positive facial expressions than men. By understanding how different groups respond to each ad, companies can create specific ads for each group.
Facial Recognition Algorithm for Increased Product Sales
Affectiva, a startup in Massachusetts, has developed software that uses facial recognition to gauge audience reactions to advertisements in real-time. This technology could revolutionize how companies assess ad effectiveness, moving away from traditional focus groups. Understanding audience emotions could lead to more targeted advertising strategies.
👥 Key Players
📰 What Happened
Affectiva has developed a facial recognition algorithm that allows companies to assess audience reactions to advertisements in real-time, moving away from traditional focus group methods. This technology analyzes facial movements to predict consumer behavior and emotional responses during ad broadcasts.
- The software uses machine learning algorithms to interpret facial movements and emotional responses.
- Affectiva's research involved monitoring over 1,200 viewers and analyzing data from over three million videos.
💡 Why It Matters
📚 Background
Facial recognition and emotion analysis technologies are becoming more prevalent in various sectors, including marketing, healthcare, and security. Understanding consumer behavior is crucial for businesses aiming to enhance their advertising strategies.
🏷️ Entities Mentioned
Translated from the original and edited for English readers. View original source →
Translation confidence: 85%