Hacking it Out: Factual at PennApps

This past weekend, I had the privilege of returning to my alma mater, the University of Pennsylvania, for the tenth installment of PennApps. PennApps is one of the country’s largest student-run hackathons— an event where students gather from around the world to spend thirty six hours coding and building new products from scratch. Partnered with Andreessen Horowitz, Factual had the opportunity to send me and my coworker to serve both as mentors and judges for the incredible student teams.

View of the project expo, where teams demoed their hacks.

It was impressive to see participants, some not yet through with high school, creating engaging games, polished apps, hacked hardware solutions, proofs of concept, and even full prototypes. Thrown in were two days of company tech talks, great local food, and all the provisions you need to live out of the engineering quad (even toothbrushes and inflatable mattresses).

In recent years, hackathons like PennApps have become a major part of the learning experience for young engineers. Students are traveling in droves to schools around the country for their chance to code, create, and compete. The culture of experimentation and taking opportunities to step away for a time and stretch oneself as a developer is one that resonates strongly at Factual (baked into our annual calendar are a series of hackathons— during these everyone from engineering to business development takes forty eight hours to focus on fun and exciting problems they don’t normally have the bandwidth to tackle). In light of this, supporting budding engineers at PennApps is just a start.

Keep an eye out for Factual mentors at these upcoming hackathons:

We can’t wait to see all of the exciting ways that students find to build products with Factual data.

Happy Hacking!
Julie Levine, Quaker, Factualite

Factual Featured Partner: SocialRadar

SocialRadar is an inventive app that combines the power of social and location data to cue users in to their connections to the people around them. This month, we had the privilege of talking with Ghafran Abbas, Chief Systems Architect, about this awesome app.

Company Name: SocialRadar
Located: Washington DC
Factual Partner Since: 2014
Website: http://www.socialradar.com/
Facebook: https://www.facebook.com/socialradarapp
Twitter: @SocialRadarApp
App Store Listing(s): iTunes, Google Play
Your Name and Title: Ghafran Abbas, Chief Systems Architect

Q: Introduce readers to SocialRadar. What does your app do?
A: When you walk into a room, enter a restaurant, attend a conference, or go to an event, SocialRadar tells you who is there, how you are connected to them, what they have been up to. By merging data from all of the top social networks (Twitter, Facebook, LinkedIn, Foursquare, Instagram, Google+) with live location information SocialRadar empowers you to make smarter real-world connections. For example, SocialRadar can tell you that there are 20 people near you whom you know – six work colleagues, four people from your college, two old high school friends, someone who plays tennis, a neighbor, and six friends-of-friends. It will tell you that one of your friends recently graduated from college and another just got a new job.

Q: Why is Factual location data important for SocialRadar?
A: We primarily use Factual’s Geopulse Geotag for reverse geocoding, so we can look up things like a user’s neighborhood, city, state, or country.

Q: What was the inspiration for SocialRadar? What problem are you trying to solve?
A: Michael Chasen, our CEO was the co-founder and CEO of Blackboard, the global e-Education company. At that time, he believed the Internet was going to fundamentally and positively change the way people teach and learn. At Blackboard, he visited hundreds of college campuses worldwide, coinciding with the rise of both social media and smartphones. He saw that the combination of mobile, social media, and location services has the potential to redefine how people meet and connect in person, so he created SocialRadar.

SocialRadar unlocks the location information from your smartphone, listens to social network chatter and gives you important knowledge about the people around you. Here are examples of how you can use SocialRadar:

  • SocialRadar provides actionable social intelligence. SocialRadar gives you real-time data on the people around you, so you can make decisions and interactions based on that information.
  • SocialRadar connects you on site with people you already know. SocialRadar is a GPS for your personal network, enabling you to quickly and effortlessly locate people everywhere from recreational gatherings such as sporting events to business functions such as trade shows. This ensures you never miss anyone nearby you know, from the former colleague two seats back whom you’ve been meaning to catch up with to the college roommate visiting your city for the weekend.
  • SocialRadar connects you to new people. SocialRadar helps you find and, if you choose, meet people with common backgrounds (same former employer, same philanthropic cause) and interests (CrossFit, crosswords). This eases social anxiety in any setting.
  • SocialRadar makes your online social networks available offline. SocialRadar provides pertinent info about the people on your social network – what they are doing real-time and where they are in the real world. This helps you overcome FOMO (Fear of Missing Out).
  • Q: Technology has made it increasingly easy to share personal thoughts and information and discover these things about others (especially mobile which allows people to publicize an idea the minute they have it). In this space, how do you walk the line between being social or informative and being “creepy?”
    A: With any new technology there will always be some that are uncomfortable but many often find it cool, especially the generations in waiting. Back in 2006, it was well publicized and viewed by many that the Facebook Newsfeed was an invasion of people’s privacy. Today, that same Newsfeed is the engine that keeps us coming back to Facebook.

    We recognize that are times where you will want to be public and times when you want to be off the grid. With SocialRadar you maintain total control over your privacy, allowing you to share your information publicly, just with your friends, be anonymous, or be invisible.

    Q: Do you find that most people prefer to be anonymous, or by virtue of being on SocialRadar are more people apt to share their locations?
    A: Since we launched back in January, we have seen about 65-70% of our users are comfortable sharing their location and keeping it on the fully public privacy setting. If you are genuinely interested in using our technology, you often realize that you can’t receive if you don’t give to the network. Meaning, if everyone is private then there really isn’t a community.

    Q: Something that really strikes me about SocialRadar is that it encourages face to face interaction (instead of trying to keep all communication digital). Where do you think the interface of social technology and live social interaction is going in the next 5-10 years?
    A: We couldn’t agree more. We always set out to change the way people meet and connect. Technology plays and will continue to play an important role in our lives. However, nothing can replace a face-to-face experience and making new memories. We have no shortage of ideas of how to encourage connections. Right now these are based off of your profile— who went to your high school or worked at your company. In the future, we want to offer more intelligent suggestions, maybe based on interests or shared experiences.

    Q: In that vein, what new or improved mobile technology would you like to see in the future? Are there any technological limitations that you face with your current product?
    A: What I would like to see more of (and I’ve sort of seen this trend) is getting better location data about people. Right now GPS does a fairly good job. But it’s limited. Things like iBeacon and Bluetooth Low Energy are helping, but it’s not there yet.

    Q: What’s a great feature about SocialRadar that users aren’t taking advantage of?
    A: Our app is pretty easy to use, but there are some “power user” features that a lot of people don’t realize exist. You can make your own connects for example, like saying “alert me when there’s an individual who went to my high school and currently lives in Washington DC within 1000 feet.” You can be very detailed about the kind of people you want to meet and be alerted when they’re nearby. We’re going to try to simplify it so more people will be able to easily set it up instead of just going with the default settings.

    Q: What’s the biggest lesson you’ve learned since starting SocialRadar?
    A: Just like many technologies, you need to crawl before you walk before you run. We know we are in the infancy stages as it pertains to geo-location and people discovery apps.

    In addition, don’t stop innovating and don’t worry about breaking things. The technology that we’ve built wouldn’t have been possible if we hadn’t had that mindset. Always try to find new things and find new ways to solve problems.

    - Julie Levine, Marketing Associate

    More About Geopulse Geotag
    Geopulse Geotag is a powerful reverse geocoder optimized for the geographic labeling of images, messages, events, and digital assets created on mobile devices. Geopulse Geotag features global coverage, high-volume API access, economical pricing, entity IDs, and an on-premise option. Learn more about Geopulse Geotag here.

    In Case You Missed It
    We’ve interviewed quite a few partners in the past few months, like Retail and Deals apps Shopular and Larky, search and discovery app Jetpac, and calling app 2GIS Dialer. See even more Featured Partners here.

Factual On Location Data, Privacy And Trust In The Mobile Space – Tyler Bell [Podcast]

Factual VP of Product, Tyler Bell joined Jacob Goldstein of Application Developers Alliance for a Voice of the Industry podcast to discuss the present and future of location data, the increasing amounts of personal information available to developers, and how to maintain the trust between developer and user while making use of that data.

Highlights include:

  • The role of location data in establishing context— the more granular the data, the more valuable.
  • Understanding general patterns with broad brushes about where the user tends to operate.
  • Privacy and the importance of respecting the bond between a user and a brand, publisher, app, etc..

Emergent Behaviors in Factual’s Geopulse Audience Profiles

Factual’s Geopulse Audience product assembles real-world profiles for millions of smart-phone users around the world. A suite of sophisticated geo-fencing, machine-learning, and heuristic methods are used to convert the user input, a set of lat/long records for a particular device, into a colorful description of the user. This description includes demographic, behavioral, and geographic information, such as a user’s age, income, ethnicity, whether they are a likely golfer, mattress shopper or electronics buyer, and which places they have visited over the past year.

As part of our ongoing QA of Geopulse Audience profiles, we had the opportunity to delve into this rich dataset of users in search of emergent consumer patterns. We asked questions such as who is likely to visit which places, are there places that are often visited in concert, and which people exhibit what consumer behaviors?

We calculated the correlation between 85 places, 9 demographic descriptors, and 25 behavioral segments for a set of more than 30 million user records making use of a Clojure library for distributed statistics.

First, we looked at a birds-eye-view of the space — the correlations for a Cartesian join on the full set of places, behavioral segments, and demographics. While at first glance this plot looks inscrutable, there are a couple of noteworthy observations. Sanity check: records are perfectly correlated with themselves. Second, there are a lot more purple squares (positive correlations) that brown squares (negative correlations). This reflects a positive bias in the way that we gather information about a particular device. An application that streams geotags more often has more geotags for ubiquitous places like Starbucks, ATMs, and movie theaters. Because the number of visits to common places is a function of the amount of geotags, place visits to Starbucks, ATMs, etc. are positively correlated. These positive correlations are more likely to surface in Audience profiles than are true negative correlations, i.e. a person who often visits McDonalds is that much more unlikely to visit Burger King. This skew toward positive correlations can also apply to behavioral segments that are learned in part based on place visits.

Also, there are several fields that look vaguely like white stripes, i.e. appear to be evenly distributed across most other fields. These white stripes include affluent consumer, age, college student, commuter, entertainment enthusiast, financial customer, female, income, leisure seeker, and NFL enthusiast. We expect income, age, and gender, to be more or less equally distributed across places because most of the places present in these records are visited by a wide range of people (McDonalds, CVS, etc.) It is also plausible that the behavioral segments like affluent consumer (one who frequents non-chain stores), financial customers (one who uses banks or ATMs), entertainment enthusiast (one who frequents movie theaters, dance clubs, etc.), date nighter (one who frequents restaurants and bars), commuter (one who travels more than five miles to get to work), leisure seeker (one who goes to playgrounds, parks, and pools), and NFL enthusiast are somewhat equally distributed across demographics, other segments, and visits to various places.

Next, we zoom in on some interesting features. We grab the 15 fields with the highest mean absolute correlation (most correlated across the board) and the 15 fields with the highest standard deviation in correlation (a gross proxy for several highly correlated fields). Birds of a feather flock together it would appear. Different car dealerships feature as a block of highly correlated fields, as car dealerships tend to be located near one another and if you’re shopping for a car you are likely to visit more than one lot. Clothing stores, such as Old Navy, JC Penney, Abercrombie and Fitch, Gap, Banana Republic, H&M, American Eagle Outfitters, and Victoria’s Secret cluster together as well, likely for the same reason (what we see in this block of highly correlated retail stores is a topology of the typical American Mall!).

Rather than beg the question whether a user is more likely to shop at two stores that are in close proximity, we can look for stores that we do not expect to be near one another but are still significantly correlated. In this case, we are interested in the elements off of the block diagonal in the above image such as the correlation between Dairy Queen and Jeep and Dodge dealers, JC Penney visitors and Mitsubishi dealers, and curious possibility that consumers are not using cash to purchase their Hyundais.

We partition places into a couple of sets (retail stores, car dealerships, ATMs/Starbucks/Movie Theaters) and identify correlations between places from different groups. We plot correlations between some of these places below. It looks like if you’re going to Napa Auto Parts you’re likely to eat burgers at Sonic, shopping at Nordstrom then you’re probably picking up pet supplies at Pet Smart, and frequenting ATM Banks in order to fuel your Starbucks addiction. In the case of the latter, this is likely a function of the fact that ATM Banks and Starbucks co-occur quite frequently (80,000 times) and just another manifestation of the more data, more visits/segments bias.

We can double check that we’ve identified the major clusters of places by running a metric multidimensional scaling algorithm on our various places (temporarily removing segments for the sake of clarity). MDS represents the distance between our various places in a 2-d space. We see that the majority of car dealerships are clustered in the top left corner, though notably Mercedes and Land Rover dealers are in their own space. In the top right hand corner, we see the Nordstrom, American Eagle, Gamestop cluster we identified earlier. We may also laugh to see the male demographic sit squarely at the center of Best Buy, Subway, and McDonalds.

As a final investigation (though with such a rich dataset there are myriad cool questions one could ask), we inquire whether particular behavioral segments are likely to co-occur and what, if any, are the demographic biases for particular consumer behavior?

A plot like the one above provides powerful insight into consumer segments. It also allows us to see whether various segments are behaving as we expect them to. Each segment was derived from an independent model, so this type of cross-examination based on external ‘truthiness’ is very powerful. For example, it seems reasonable that a user is less likely to be a college student the older they are or the more they appear to be a frequent traveller. Affluent consumers have higher incomes, date nighters overlap a fair bit with entertainment enthusiasts, and males are more likely than females to be golfers.

This exploration is the first of a series in which we QA Geopulse Audience profiles by comparing what they encode about user segments, demographics, and place visits to what we know about those fields based on external sources of information, such as consumer surveys and census data. These analyses enable us to refine our models and maximize the quality of our Geopulse Audience product.

Please email me at natasha@factual.com if you have any questions or feedback about these results or if you would like to learn more about being a Data Engineer at Factual!

- Natasha Whitney, Data Engineering Intern

1. In order to ensure that our results will hold for samples of varying sizes, we only included correlations that had a 5% or smaller probability of occurring by mere chance given the number of records that were used to calculate the correlation (see this discussion of statistical significance for correlations).

InMobi Inks Global Partnership with Factual to Increase Mobile Ad Efficacy for Brands

Today we announced a partnership with InMobi, the world’s largest independent mobile advertising platform, to create a rich set of audience targeting solutions for brand advertisers. InMobi’s anonymous data set of global mobile consumer activity will be combined with Factual’s location-based data to build geo-derived audience segments worldwide. This partnership enables marketers to globally leverage a richer set of geo-targeting features that incorporate geo-based consumer intelligence at a large scale.

Brands are often challenged by ‘ad spillage,’ or dollars wasted when an advertisement reaches audiences outside their geo-spatial target segment. Providing advertisers with rich contextual data on consumers enables delivery of relevant ads with higher campaign efficacy, or lower ad spillage.

InMobi gathers anonymous user data from the 759 million monthly active unique users on its network. To enhance an advertiser’s targeting efficacy, InMobi builds on this understanding of users by partnering with multiple data-specialists across the ecosystem, such as Factual.

The partnership between InMobi and Factual will help advertisers understand how consumers move through the physical world by using aggregated anonymous user location data. Factual’s Global Places data covers over 65 million businesses and points of interest across 50 countries. Geopulse Audience uses this understanding of geography to build location-based profiles that contain hundreds of non-private behavioral attributes describing a user. For example, brands can now run campaigns targeting college students at Starbucks coffee shops or live sports fans at a stadium. In addition, InMobi will provide brands with geographically differentiated insights that will enable marketers to build targeted campaigns. For instance, brands will be able to market to movie goers in New York differently than to movie goers in London.

“As consumers spend more time on their mobile phones, our investments in building the infrastructure that gleans powerful data signals and insights from these interactions will provide increasing value to marketers,” said Anne Frisbie, Vice President & General Manager – Global Alliances, InMobi. “We are committed to enabling consumers and businesses to make smarter decisions. By partnering with Factual, we are able to improve our geo-targeting capabilities, thereby offering more relevant and engaging mobile marketing. Of equal importance, we are able to offer targeted audience buying to brands not just in the US, but around the world.”

Every market in the world is being transformed by mobile – the valuable signals generated by mobile devices enable marketers to deliver contextually relevant user experiences that were previously impossible. We believe that advertisers and developers everywhere are going to benefit from the combination of InMobi’s global reach and the scale of its mobile consumer data, combined with our global location data and understanding.

-Bill Michels, SVP Product and Partnerships