Private:progress-neshat

2011-01-25T00:36:10Z

Hsadeghi:

2011-01-10T20:38:46Z

Advertising in online videos is a large and
growing market. In this paper, we propose a new approach
to match ads with online videos based on the shopping
interests of the target audience of videos. The proposed
approach increases the relevance of ads to the actual
viewers (humans) of videos, which increases the number
of users who purchase goods and services offered by
the advertisers. This in turn will increase the revenues
for advertisers as well as the video sites as video sites
usually charge advertisers based on the number of user
clicks on their ads. The proposed approach is different
from current approaches used in practice or proposed in
the literatures, which most of them try to maximize the
relevance of ads to the tags or contents of videos (objects).
We conduct a subjective study to evaluate the performance
of the proposed approach on many videos retrieved from
YouTube. Our results show that the proposed approach
yields more relevant ads to the viewers than the YouTube’s
approach. We also compare against other approaches
proposed in the literature and we show that the new
approach outperforms them.
== People ==

* [http://www.cs.sfu.ca/~mhefeeda/ Mohamed Hefeeda]

* [http://www.cs.sfu.ca/~hsadeghi/ Hamed Sadeghi Neshat (MSc student)]

== On-going Research Problems ==

Estimating the click-through rate for new ads with semantic and feature based similarity algorithms

== References and Links ==

Network and Multimedia Systems Lab (NMSL)

2011-01-10T20:37:31Z

Hsadeghi:

__NOTOC__

'''Welcome to the Network Systems Lab (NSL) at SFU!'''

We are interested in the broad areas of computer networking and multimedia systems. We develop algorithms and protocols to enhance the performance of networks, especially the Internet, and to efficiently distribute multimedia content (e.g., video and audio objects) to large-scale user communities. The Network Systems Lab is led by [http://www.cs.sfu.ca/~mhefeeda/ Dr. Mohamed Hefeeda], and is affiliated with the [http://www.cs.sfu.ca/research/groups/NML/ Network Modeling Group] at SFU.
The NSL lab is located in room SUR 4120 (Surrey campus).

We hold regular [[group meeting]] for discussion and brainstorming.

Our current research interests include multimedia networking, peer-to-peer systems, wireless sensor networks, and network security. Brief description and links to currently active projects are given below.

== '''Peer-to-Peer Systems''' ==

We are exploring the applicability of the P2P paradigm to build cost-effective content distribution systems. Problems such as sender selection, adaptive object replication, and content caching are being studied. We are also developing models to analyze the new characteristics of the P2P traffic and the impact of these characteristics on the cache replacement policies and object replication strategies.
Furthermore, we are devising analytic models to study the dynamics of the P2P system capacity and the impact of various parameters on it.

* '''[[pCDN|pCDN: Peer-assisted Content Distribution Network]]'''

* '''[[Modeling and Caching of P2P Traffic]]'''

== '''Multimedia Networking''' ==

We are focusing on distributed streaming in dynamic environments and for heterogeneous clients. Our
goal is to analyze and understand scalable coding techniques, and to design several optimization and streaming algorithms to make the best possible use of them in real multimedia systems. This will yield better quality for users, and more efficient utilization of network and server resources. We are also designing algorithms to optimize streaming quality for wireless and mobile clients.

* '''[[Scalable Multimedia Streaming]]'''

* '''[[mobileTV|Mobile TV Networks]]'''

* '''[[wimax|Multimedia Streaming in WiMAX Networks]]'''

* '''[[CanVid|CanVid: Content- and Network-aware Video Processing]]'''

== '''Online Networked Games''' ==

We are designing various algorithms to improve the performance of online games.

* '''[[Minimizing Round-Trip Time in Online Games]]'''

* '''[[Minimizing Energy Consumption for Online Games on Mobile Phones]]'''

== '''High Performance Computing''' ==

We are exploring the opportunities of utilizing new architectures such as GPUs, multi-core processors, and distributed clusters (cloud computing) to efficiently solve research problems related to multimedia content analysis,
large-scale data analysis, and machine learning techniques.

* '''[[hpc|Approximation algorithms for Kernel Methods on Multi-core CPUs and GPUs]]'''

* '''[[Accelerating Online Auctions with parallel implementation on GPU]]'''

* '''[[videoInfringement | Video Copy Detection using Cloud Computing]]'''

== '''Online advertising''' ==
Online advertising is a form of promotion that uses the Internet and World Wide Web for advertising a good or service to attract users. Examples of online advertising include contextual ads on search engine results pages,, banner ads, Rich Media Ads.

* '''[[Predicting ads' quality ]]'''

* '''[[SmartAd: A Smart System for Effective Advertising in Online Videos ]]'''

= '''''Concluded Projects''''' =

== '''Wireless Sensor Networks''' ==

We are developing coverage and connectivity maintenance protocols that consider probabilistic (i.e., more realistic) sensing and communication models. We are also designing protocols that provide controllable degrees of coverage (k-coverage).

* '''[[Probabilistic Coverage and Connectivity]]'''

* '''[[K-Coverage and its Application to Forest Fire Detection]]'''

== '''Network Security''' ==

We are exploring network monitoring techniques to detect and thwart intrusion and denial-of-service attacks in their early stages by observing unusual traffic patterns injected by such attacks. We are studying the security of multimedia streaming systems that employ multi-layer and fine-grain scalable video streams.

* '''[[Security of the SIP protocol]]'''

* '''[[Detecting DoS Attacks and Service Violations in QoS-enabled Networks]]'''

* '''[[Security of Scalable Multimedia Streams]]'''

Predicting ads' quality

2010-11-19T19:31:45Z

Hsadeghi:

Predicting ads' quality

2010-11-19T19:28:13Z

Hsadeghi:

Internet advertising is the main source of income for search engines today. As the number of Internet users increases, the Internet advertising becomes increasingly popular among people who want to advertise a service or a product. Google reported $6,475 million revenue from advertisement in 2009 which is 8% more than the previous year. This, emphasizes the fact that internet advertising is a widely attractive and growing market for advertisers and search engines.

When a user enters a query in a search engine, there are often some sponsored links or ads presented alongside with the search results. These ads are chosen by an auction between all candidate ads which have keywords similar to the user entered query. In this auction, winners will be chosen based on two factors: offered bid and quality. In this article, quality means the ability to attract more users' clicks. Advertisers usually want to place their ads in the best spot in the page without paying more money, so they try to increase the quality of ads by choosing good terms for title and descriptions. On the other side, Search engines use the Price Per Click (PPC) model for Internet advertising. In this model search engines can earn money just if somebody clicks on the displayed ads and as a result, there is no cost for the advertisers merely because of ad appearance. So for earning maximum revenue, search engines also try to select ads with better quality to attract more clicks. Roughly speaking, ads with high quality are important for both advertiser and search engine.

For the ads which have been in the system for longer periods of the time, we can find their quality just by looking at their click through rate (CTR). If an ad had higher amount of CTR, it is more attractive to users and has better quality. But for new ads or for those ones without enough historical data, we should find another way to estimate their quality.

== People ==

* [http://www.cs.sfu.ca/~mhefeeda/ Mohamed Hefeeda]

* [http://www.cs.sfu.ca/~hsadeghi/ Hamed Sadeghi Neshat (MSc student)]

== On-going Research Problems ==

The propose of current study is to investigate on a quality measurement method which works with conceptual and feature based similarity algorithms. The proposed approach can find similar ads from historical data and estimate ads’ quality for new ads in compare with current ads in the Internet. More over, in this experiment we will examine some novel machine learning methods and use them in the Internet and advertising concepts and try to customize them in order to be effective in these areas.

== References and Links ==