Cultural Analytics

Due to big data’s volume and variety, coding proficiency and machine learning tools become part and parcel of the new humanities. Every week 28 billion photos are uploaded to Google Photos. YouTube is a repository of an estimated 95,000 years of video. Cultural analytics is a necessity for navigating binary large objects of this large scale.

Big Data Studies Lab pursues the following experiments in cultural analytics. Our goal is not necessarily to identify patterns or provide statistical summaries of cultural data sets, but to create intuitive tools that help humanists discover the equivalent of “walking through the stacks” in an age of big data.

• a browser for user-downloaded social media archives
• auto-generation of frame samples along with object and emotion detection using computer vision
• critical evaluation of machine learning models used for gender and emotion classification
• music database APIs