Online Advertising Ecosystem / Network data

Начать. Это бесплатно
или регистрация c помощью Вашего email-адреса
Online Advertising Ecosystem / Network data создатель Mind Map: Online Advertising Ecosystem / Network data

1. User privacy

1.1. Privacy leakages

1.1.1. Complex problem (requires monitoring all the communications)

1.1.1.1. Active measurements

1.1.1.1.1. Nikos/Ruben TYPES Focus on detecting OBA

1.1.1.1.2. Northeastern guys

1.1.1.2. Passive mearurements

1.1.1.2.1. Haystack (Narseo)

1.1.1.2.2. ReCon (Choffness)

1.2. Online tracking

1.2.1. Tracker identifying

1.2.1.1. Detecting and Defending Against Third-Party Tracking on the Web NSDI'12

1.2.1.2. Anatomy of the Third-Party Web Tracking Ecosystem

1.2.1.3. The Web never forgets: Persistent tracking mechanisms in the wild

1.2.1.4. Hassan/Mellia Unsupervised Detection of Web Trackers

1.2.1.5. Adblock plus, ghostery, blur, privacybadge...

1.2.1.6. TrackAdvisor

1.2.2. Blocking/detecting IDs

1.2.2.1. Cookies

1.2.2.1.1. Cookie Mining (Cookie jar)

1.2.2.1.2. Hassan/Mellia (it fails)

1.2.2.1.3. Nikos (WIT)  (Not applicable, I hesitate if it really protect the users)

1.2.2.2. Others

1.2.2.2.1. INRIA MyTrackingChoices https://myrealonlinechoices.inrialpes.fr/

1.2.2.2.2. Cliqz (blocks all third party cookies, blocks parts of the url request that make users unique)

1.2.2.2.3. Towards Mining Latent Client Identifiers from Network Traffic

1.2.2.3. Open challenges

1.2.2.3.1. Correctly detect user IDs

1.2.3. Detecting cookie matching??

1.2.4. How to avoid it??

1.2.4.1. Adding "Noise" to the profiles

1.2.4.2. Web-page level sandboxing

2. Data monetization Network services

2.1. User profiling

2.1.1. Assign keywords to pages

2.1.1.1. How is Google doing it? can we compare using adwords?

2.1.1.2. Mathias ??

2.1.1.3. Natural language processing??

2.1.2. What is a user in the network??

2.2. How to do it over encrypted traffic?

2.2.1. Can we build the system?? (working with Telefonica)

2.2.1.1. Make the training scalable

2.2.1.1.1. Data collection

2.2.1.1.2. Classifier training

2.2.1.2. Make the identification fast

2.2.1.3. How easy is to identify when a user has gone to a new page

2.2.1.4. How may classifier per page (different browser/devices have different behaviour?)

2.2.2. Can we improve the results?

2.2.2.1. Machine learning techinque

2.2.2.2. Signature representation

2.2.3. Trusted proxies??

2.2.3.1. Users will reject the idea

2.2.3.1.1. https://blog.avast.com/2015/05/25/explaining-avasts-https-scanning-feature/

2.2.3.1.2. http://arstechnica.com/security/2015/02/lenovo-pcs-ship-with-man-in-the-middle-adware-that-breaks-https-connections/

2.2.3.2. public key pinning

2.2.3.2.1. https://blog.mozilla.org/security/2014/09/02/public-key-pinning/

2.2.3.2.2. https://www.owasp.org/index.php/Certificate_and_Public_Key_Pinning

2.2.3.3. Are ISPs already doing it??

2.2.3.3.1. A Tangled Mass: The Android Root Certificate Stores

2.2.4. BlindBox: Deep Packet Inspection over Encrypted Traffic

2.2.5. Other technique to be replicated over encrypted traffic??

2.3. User value

2.3.1. How valuable is a user?

2.3.2. What is the value of the information?

2.4. Privacy preserving solutions

2.4.1. Paladin

2.4.2. Data Anonymization

2.4.2.1. We are not experts on this

2.5. Modifying packets on the fly??

2.6. Content prioritation??

2.6.1. Load ads at the end

2.6.1.1. Implications?

2.6.1.2. RTB confirmation?

2.7. Ads on IPTV

2.7.1. NEC provides the TV system for pccw http://video-streaming.orange.fr/autres/nec-onlinetv-iptv-solution-VID0000001LUX9.html

2.7.1.1. http://www.showcase-nec.com/iptv-delivery-high-resolution-video-wide-area-networks/

2.8. User tracking

2.8.1. Cross devices linking

2.8.1.1. http://adexchanger.com/data-exchanges/a-marketers-guide-to-cross-device-identity/

2.8.1.2. http://www.verizonwireless.com/featured/precision/

2.8.2. Verizon ads a supercookie to all the communications

2.8.2.1. http://www.verizonwireless.com/support/relevant-mobile-ad/

2.8.2.2. https://www.eff.org/es/deeplinks/2014/11/verizon-x-uidh

2.8.3. AT&T

2.8.3.1. http://www.att.com/gen/press-room?pid=22665&cdvn=news&newsarticleid=34198&mapcode=

2.8.3.2. http://adworks.att.com/targeting.html?lpos=Header:2

3. RTB techniques

3.1. How to serve better ads

3.1.1. Old topic

3.1.2. Available data http://contest.ipinyou.com/

3.1.3. good machine learning problem

4. GDPR https://en.wikipedia.org/wiki/General_Data_Protection_Regulation

4.1. How it affect to the nowadays business

4.1.1. Paladin

4.1.2. Can we identify things that will become ilegal?

4.1.3. It should bring business opportunities