--======== Review Reports ========--

The review report from reviewer #1:

*1: Is the paper relevant to WI?
[_] No
[X] Yes

*2: How innovative is the paper?
[_] 5 (Very innovative)
[X] 4 (Innovative)
[_] 3 (Marginally)
[_] 2 (Not very much)
[_] 1 (Not)
[_] 0 (Not at all)

*3: How would you rate the technical quality of the paper?
[_] 5 (Very high)
[_] 4 (High)
[X] 3 (Good)
[_] 2 (Needs improvement)
[_] 1 (Low)
[_] 0 (Very low)

*4: How is the presentation?
[_] 5 (Excellent)
[_] 4 (Good)
[X] 3 (Above average)
[_] 2 (Below average)
[_] 1 (Fair)
[_] 0 (Poor)

*5: Is the paper of interest to WI users and practitioners?
[X] 3 (Yes)
[_] 2 (May be)
[_] 1 (No)
[_] 0 (Not applicable)

*6: What is your confidence in your review of this paper?
[_] 2 (High)
[X] 1 (Medium)
[_] 0 (Low)

*7: Overall recommendation
[_] 5 (Strong Accept: top quality)
[X] 4 (Accept: a regular paper)
[_] 3 (Weak Accept: could be a poster or a short paper)
[_] 2 (Weak Reject: don't like it, but won't argue to reject it)
[_] 1 (Reject: will argue to reject it)
[_] 0 (Strong Reject: hopeless)

*8: Detailed comments for the authors
The paper presents an approach for mining information about user migration among Web sites with the purpose of analyzing crowds' interests when navigating over inter-connected Web pages. The authors argue that such a kind of information mining, which aims at building a "user-driven" Web network, is useful to applications that need to find out the most relevant Web pages to be searched in a Web site.

In the following, positive aspects and suggestions for paper improvement are given.

Positive aspects
----------------
- a methodology for obtaining a user-driven Web network based on Web crawling and generation of a Web graph where nodes are Web pages and weighted edges that denotes how many common users visit both pages.
- an approach to normalize edge weights in different scales as well as to generate an absolute weight.
- an analysis of some relevant network properties, specially the seed-free and scale-independent phenomena.
- the development of a "Website selector" tool that shows the applicability of the proposed methodology in an advertinsing domain.


Suggestions for Paper Improvement
---------------------------------
- Paper is redundant when describing contributions. On the other hand, motivation about the originality of the approach when compared to related work is weak. There are few Computer Science references regarding Web Mining area, which seems that a more deep investigation about the theme could be accomplished.
- A concrete example of the network normalization process could be presented for sake of understanding. It is clear that the focus is on the quality of the normalized network generation, but there is no concern about performance (several days for analyzing 3 sets of ~300K pages). Immediate future work should consider this, given the focus on decision making applications.
- Some of the comparison between a user-driven network and the Web network seems to be obvious, specially when it is said that Web nodes are less connected (Average path lenght and Clustering Coefficient results).
- References about non-linear optimization and revenue models could be provided.



========================================================
The review report from reviewer #2:

*1: Is the paper relevant to WI?
[_] No
[X] Yes

*2: How innovative is the paper?
[_] 5 (Very innovative)
[X] 4 (Innovative)
[_] 3 (Marginally)
[_] 2 (Not very much)
[_] 1 (Not)
[_] 0 (Not at all)

*3: How would you rate the technical quality of the paper?
[_] 5 (Very high)
[X] 4 (High)
[_] 3 (Good)
[_] 2 (Needs improvement)
[_] 1 (Low)
[_] 0 (Very low)

*4: How is the presentation?
[_] 5 (Excellent)
[X] 4 (Good)
[_] 3 (Above average)
[_] 2 (Below average)
[_] 1 (Fair)
[_] 0 (Poor)

*5: Is the paper of interest to WI users and practitioners?
[X] 3 (Yes)
[_] 2 (May be)
[_] 1 (No)
[_] 0 (Not applicable)

*6: What is your confidence in your review of this paper?
[X] 2 (High)
[_] 1 (Medium)
[_] 0 (Low)

*7: Overall recommendation
[_] 5 (Strong Accept: top quality)
[X] 4 (Accept: a regular paper)
[_] 3 (Weak Accept: could be a poster or a short paper)
[_] 2 (Weak Reject: don't like it, but won't argue to reject it)
[_] 1 (Reject: will argue to reject it)
[_] 0 (Strong Reject: hopeless)

*8: Detailed comments for the authors
This is a good paper. I have learned a lot from it. Its first contribution is to have inferred the co-view relationship between websites from Google trends. Its second contribution is to have shown the merit of putting co-view relationship of websites into Internet AD planning.

However, I feel the paper's title strange. The paper title doesn't reflect the actual content of the paper. Specifically, the co-view relationship inferred from Google trends didn't reveal any information about "crowd migration," which should be referred to a long-term movement from website A to another website B. Instead, Google trends' data merely show that a user visits A also visits B. The authors are strongly suggested to change the paper's title to reflect its content and contributions.


========================================================
The review report from reviewer #3:

*1: Is the paper relevant to WI?
[_] No
[X] Yes

*2: How innovative is the paper?
[_] 5 (Very innovative)
[_] 4 (Innovative)
[X] 3 (Marginally)
[_] 2 (Not very much)
[_] 1 (Not)
[_] 0 (Not at all)

*3: How would you rate the technical quality of the paper?
[_] 5 (Very high)
[_] 4 (High)
[X] 3 (Good)
[_] 2 (Needs improvement)
[_] 1 (Low)
[_] 0 (Very low)

*4: How is the presentation?
[_] 5 (Excellent)
[_] 4 (Good)
[X] 3 (Above average)
[_] 2 (Below average)
[_] 1 (Fair)
[_] 0 (Poor)

*5: Is the paper of interest to WI users and practitioners?
[X] 3 (Yes)
[_] 2 (May be)
[_] 1 (No)
[_] 0 (Not applicable)

*6: What is your confidence in your review of this paper?
[_] 2 (High)
[X] 1 (Medium)
[_] 0 (Low)

*7: Overall recommendation
[_] 5 (Strong Accept: top quality)
[X] 4 (Accept: a regular paper)
[_] 3 (Weak Accept: could be a poster or a short paper)
[_] 2 (Weak Reject: don't like it, but won't argue to reject it)
[_] 1 (Reject: will argue to reject it)
[_] 0 (Strong Reject: hopeless)

*8: Detailed comments for the authors
The paper presents a method to derive the average number of users that visits pairs of web sites over longer timescales. This is used to build a weighted graph whose nodes are websites: the weigh associated with edges represents number of shared visiting users. The properties of such a graph are studied (in different domains) and one application with experimental results is presented.
Overall the paper is well written and results are clear.

========================================================
The review report from reviewer #4:

*1: Is the paper relevant to WI?
[_] No
[X] Yes

*2: How innovative is the paper?
[_] 5 (Very innovative)
[X] 4 (Innovative)
[_] 3 (Marginally)
[_] 2 (Not very much)
[_] 1 (Not)
[_] 0 (Not at all)

*3: How would you rate the technical quality of the paper?
[_] 5 (Very high)
[X] 4 (High)
[_] 3 (Good)
[_] 2 (Needs improvement)
[_] 1 (Low)
[_] 0 (Very low)

*4: How is the presentation?
[_] 5 (Excellent)
[X] 4 (Good)
[_] 3 (Above average)
[_] 2 (Below average)
[_] 1 (Fair)
[_] 0 (Poor)

*5: Is the paper of interest to WI users and practitioners?
[X] 3 (Yes)
[_] 2 (May be)
[_] 1 (No)
[_] 0 (Not applicable)

*6: What is your confidence in your review of this paper?
[X] 2 (High)
[_] 1 (Medium)
[_] 0 (Low)

*7: Overall recommendation
[_] 5 (Strong Accept: top quality)
[X] 4 (Accept: a regular paper)
[_] 3 (Weak Accept: could be a poster or a short paper)
[_] 2 (Weak Reject: don't like it, but won't argue to reject it)
[_] 1 (Reject: will argue to reject it)
[_] 0 (Strong Reject: hopeless)

*8: Detailed comments for the authors
The paper provides an interesting und reasonably deep study into migration patterns on the Web as they pertain to users or small collections of users. The findings are definitely relevant to advertisers, as they show how ad visibility can be increased significantly. To achieve this result, the authors present detailed algorithm descriptions based on a graph representation on the Web as well as on network properties.

My only (minor) criticism is that the authors could (somewhere in the paper) say a word about whether everything they propose is efficient, i.e., of reasonable (low) complexity.

========================================================