librec_trust — FilmTrust and CiaoDVD trust and rating networks (2011-2013)

Description

Trust networks and ratings of users of two movie websites, crawled by G. Guo and collaborators for research on trust-aware recommender systems, and distributed with the LibRec library:

• FilmTrust, a movie sharing and rating website, crawled in its entirety in June 2011. Users rate movies, and can declare that they trust other users.
• CiaoDVD, the DVD category of the British consumer review website Ciao (dvd.ciao.co.uk), crawled in November and December 2013. Users write reviews of movies with a rating, rate the helpfulness of other users' reviews, and declare which other users they trust. Movies without any review were left out by the authors.

User and movie identifiers were anonymized; the identifiers of users are the same in the networks of one website. There are five networks:

• filmtrust_trust: a directed network of users, where an edge $i \to j$ means that user $i$ trusts user $j$.
• filmtrust_ratings: a bipartite network of users and the movies they rated, with the rating. Three user-movie pairs appear twice in the data (with the same or a different rating); all rows are kept as edges.
• ciaodvd_trust: a directed network of users, where an edge $i \to j$ means that user $i$ trusts user $j$.
• ciaodvd_ratings: a bipartite network of users and the movies they reviewed, with the rating and the date of the review. A user may have reviewed a movie more than once (320 user-movie pairs appear more than once); each review is an edge.
• ciaodvd_review_ratings: a directed multigraph of users, where an edge $i \to j$ means that user $i$ rated the helpfulness of a review written by user $j$ (the author of the review is given by ciaodvd_ratings). Each rating is an edge. Users who did not write reviews or rate them appear only in the trust network.

The trust values given in the data are all 1, and are therefore left out.

The networks have the following properties:

• user_id (users), movie_id (movies): the anonymized identifiers in the data; is_movie (bipartite networks), whether the node is a movie (otherwise a user).
• ciaodvd_ratings, vertex genre: the numeric identifier of the movie's genre (one of 17; their names are not given, -1 for users).
• filmtrust_ratings, edge rating: the rating of the movie (from 0.5 to 4, in steps of 0.5).
• ciaodvd_ratings, edges: rating, the rating of the movie (1 to 5); review_id, the identifier of the review; time, the date of the review, as the Unix time of 00:00 UTC of that day.
• ciaodvd_review_ratings, edges: review_id, the identifier of the rated review; helpfulness, the rating of the review: 0 (off topic), 1 (not helpful), 2 (somewhat helpful), 3 (helpful), 4 (very helpful) or 5 (exceptional).

Tags

Social Online Preferences Weighted Multigraph Timestamps Metadata

Citation

• G. Guo, J. Zhang, and N. Yorke-Smith, "A Novel Bayesian Similarity Measure for Recommender Systems." Proceedings of the 23rd International Joint Conference on Artificial Intelligence (IJCAI), 2619-2625 (2013). www.ijcai.org
• G. Guo, J. Zhang, D. Thalmann, and N. Yorke-Smith, "ETAF: An Extended Trust Antecedents Framework for Trust Prediction." Proceedings of the 2014 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), 540-547 (2014). 10.1109/ASONAM.2014.6921639 Sci-Hub

Upstream URL OK

https://guoguibing.github.io/librec/datasets.html

Networks

Tip: click on a column header to sort the list, or hover over it for a description.

Column legend

$\left<k\right>$

Mean degree

$\sigma_k$

Standard deviation of the degrees

$\lambda_h$

Largest eigenvalue of the Hashimoto (non-backtracking) matrix

$\tau$

Random walk mixing time

$r$

Degree assortativity

$c$

Global clustering coefficient

$\oslash$

(Pseudo-) diameter

$S$

Relative size of the largest component

NPs

Node properties

EPs

Edge properties

Show all columns

Name	Nodes	Edges	$\left<k\right>$	$\sigma_k$	$\lambda_h$	$\tau$	$r$	$c$	$\oslash$	$S$	Kind	Mode	NPs	EPs	gt	GraphML	GML	csv
filmtrust_trust	874	1,853	2.12	7.40	12.38	58.58	0.06	0.20	13	0.70	Directed	Unipartite	user_id		19 KiB	26 KiB	26 KiB	22 KiB
filmtrust_ratings	3,579	35,497	19.84	71.55	140.13	6.20	-0.59	0.00	7	1.00	Undirected	Bipartite	is_movie user_id movie_id	rating	107 KiB	231 KiB	227 KiB	168 KiB
ciaodvd_trust	4,658	40,133	8.62	39.72	90.05	15.34	0.18	0.21	9	0.98	Directed	Unipartite	user_id		148 KiB	239 KiB	237 KiB	189 KiB
ciaodvd_ratings	33,736	72,665	4.31	16.72	34.97	101.63	-0.13	0.00	19	0.95	Undirected	Bipartite	is_movie user_id movie_id genre	rating review_id time	956 KiB	1.4 MiB	1.4 MiB	1.3 MiB
ciaodvd_review_ratings	30,185	1,625,480	53.85	848.71	369.19	29.63	0.00	0.50	10	1.00	Directed	Unipartite	user_id	review_id helpfulness	5.1 MiB	10.3 MiB	10.1 MiB	8.7 MiB

Ridiculograms^*

^* These are automatically generated force-directed visualizations, and can be quite meaningless for networks both big and small. They should not be taken seriously as sources of scientific insight. See here for a discussion.