background preloader

GazeHawk - Webcam Eye Tracking - Ad Testing and Optimization

GazeHawk - Webcam Eye Tracking - Ad Testing and Optimization
Related:  Computer Vision

Predict Gaze Marin Software Integrates Facebook Ads Into Paid Search Management Platform Designing for the Mobile Web: Special Considerations By Shanshan Ma Published: January 17, 2011 “In comparing the design of mobile Web sites with the design of Web sites for computers, I realized that complex context is another important factor that differentiates the two platforms.” There are several differences between designing a Web site for a computer and designing one for a smartphone. In addition to the four problems Nielsen wrote about, I’ll cover design for complex contexts of use in my discussion of constraints on mobile Web sites. Small Screens Some ways of coping with small screens include prioritizing features and content reducing levels of hierarchy for content communicating workflows Prioritizing Features and Content “It’s usually necessary to sacrifice some of a Web site’s features and content when designing a mobile version of the site, so users can easily find the ones they really need.” Current Web design best practices and guidelines assume a 1024x768 screen resolution. Figure 1—United mobile home page Communicating Workflows

LinkedIn Launches New Analytics For Company Pages | B2C Marketing Insider LinkedIn has always been the job seekers platform. Looking for a job, look no further than LinkedIn. While they have added features to enhance their “stay power” with Twitter updates, the LinkedIn sharing tool, the ability to upgrade to place your resume high atop of a pile, to our updates and groups, their focus has always been to be a professional site where users can come and be professional and build a platform that was conducive to connecting with people to gain employment or build a business. Their new analytics for company pages have turned the tables a bit so that a company can see where their presence is being recognized. LinkedIn’s New Analytics for Company Pages LinkedIn has launched three new analytics for companies. 1. 2. 3.

Gender Classification with OpenCV — OpenCV v2.4.9 documentation Introduction A lot of people interested in face recognition, also want to know how to perform image classification tasks like: Gender Classification (Gender Detection)Emotion Classification (Emotion Detection)Glasses Classification (Glasses Detection)... This is has become very, very easy with the new FaceRecognizer class. Prerequisites For gender classification of faces, you’ll need some images of male and female faces first. Angelina JolieArnold SchwarzeneggerBrad PittEmma WatsonGeorge ClooneyJennifer LopezJohnny DeppJustin TimberlakeKaty PerryKeanu ReevesNaomi WattsPatrick StewartTom Cruise Once you have acquired some images, you’ll need to read them. Let’s dissect the line. /home/philipp/facerec/data/gender/male/keanu_reeves/keanu_reeves_01.jpg;0 /home/philipp/facerec/data/gender/male/keanu_reeves/keanu_reeves_02.jpg;0 /home/philipp/facerec/data/gender/male/keanu_reeves/keanu_reeves_03.jpg;0 ... All images for this example were chosen to have a frontal face perspective. Running the Demo

Federated Media Publishing The Web Analyst's Code of Ethics - Web Analytics Association The Digital Analytics Association exists to help organizations illuminate and overcome the challenges of data acquisition, exploration, deduction and application. The DAA is a not-for-profit, volunteer-powered association, and strives to help individuals become more valuable through education, community, research and advocacy. The science of analysis using data to understand historical patterns with an eye to improving performance and predicting the future. The analysis of digital data refers to information collected in interactive channels (online, mobile, social, etc.). Digital Analytics has become an integral part of core business strategies and maintaining a competitive edge. Digital data started the Big Data meme as it heralded the onslaught of Volume, Variety and Velocity, opening the door to new types of correlative discovery much wider.

Features - Mixpanel | Real-time Web Analytics, Funnel Analysis There are many metrics that measure engagement but page views are not one of them. Make your product better by measuring actions, instead of page views. Mixpanel gives you the ability to easily measure what people are doing in your app on iOS, Android, and web. See how it works Answer your hardest questions without writing SQL Working with data often means writing SQL queries, which is great for engineers or data scientists but hard or impossible for everyone else. “An overlooked value of @mixpanel is that the non-technical folks can answer their own questions about your data.” Eric Normand Point and click analytics for mobile Understanding what your users are doing doesn’t have to be a complicated setup process. Features Drill into your data with segmentation so you can find out where to focus your energy when building your product. Visualize your data in different ways. Bookmarks let you save reports you look at a lot so you can save time. Getting started takes less than 10 minutes

blog:gender_classification [ My last post was very long and I promise to keep this one short. In this post I want to do gender classification on a set of face images and find out which are the specific features faces differ in. Dataset I have used the celebrity dataset from my last post and downloaded images for Emma Watson, Naomi Watts and Jennifer Lopez. So the database has 8 male and 5 female subjects, each with 10 images. All images were chosen to have a frontal face perspective and have been cropped to a size of 140x140 pixels, just like this set of images. Experiments Get the code from github by either cloning the repository: $ git clone git@github.com:bytefish/facerec.git or downloading the latest version as a tar or zip archive. Then startup Octave and make the functions available: $ cd facerec/m $ octave octave> addpath(genpath(".")); octave> fun_fisherface = @(X,y) fisherfaces(X,y); % no parameters needed octave> fun_predict = @(model,Xtest) fisherfaces_predict(model,Xtest,1); % 1-NN Fisherface and Conclusion

Federated Media Buys Semantic Profiling Technology From TextDigger Ad network Federated Media is announcing an acquisition today, buying a semantic profiling technology platform from semantic search startup TextDigger today. Terms of the deal were not disclosed. TextDigger, which was incubated at CNET in 2005, spun out a platform for semantic search and related services. TextDigger’s technology allows owners of large content collections to add semantics indexing on top of an existing keyword search engine. Federated Media will use the new technology to add semantic technology to content tagging, filtering, topic extraction, and SEO. The technology will also be used for ad targeting, and semantic search engine optimization for a site or network of sites. Tim Musgrove, TextDigger’s founder, will join Federated as Chief Scientist, while retaining an affiliation with TextDigger as their Senior Research Fellow. After shopping the company around in 2008 and not finding a buyer willing to pay his price, Federated raised a $50 million investment round instead.

Features - Mixpanel | Real-time Web Analytics, Funnel Analysis There are many metrics that measure engagement but page views are not one of them. Make your product better by measuring actions, instead of page views. Mixpanel gives you the ability to easily measure what people are doing in your app on iOS, Android, and web. See how it works Answer your hardest questions without writing SQL Working with data often means writing SQL queries, which is great for engineers or data scientists but hard or impossible for everyone else. “An overlooked value of @mixpanel is that the non-technical folks can answer their own questions about your data.” Eric Normand Point and click analytics for mobile Understanding what your users are doing doesn’t have to be a complicated setup process. Features Drill into your data with segmentation so you can find out where to focus your energy when building your product. Visualize your data in different ways. Bookmarks let you save reports you look at a lot so you can save time. Getting started takes less than 10 minutes

Twitalyzer: Serious Analytics for Social Relationships

Related: