-
公开(公告)号：US20160070709A1
公开(公告)日：2016-03-10
申请号：US14849227
申请日：2015-09-09
Applicant: STC.UNM
Inventor： Shuang Luan , Abdullah Mueen , Michalis Faloutsos , Amanda J. Minnich
IPC: G06F17/30
CPC classification number: G06Q30/0282
Abstract: Multiple sources of reviews for the same product or service (e.g. hotels, restaurants, clinics, hair saloon, etc.) are utilized to provide a trustworthiness score. Such a score can clearly identify hotels with evidence of review manipulation, omission and fakery and provide the user with a comprehensive understanding of the reviews of a product or establishment. Three types of information are used in computing the score: spatial, temporal and network or graph-based. The information is blended to produce a representative set of features that can reliably produce the trustworthiness score. The invention is self-adapting to new reviews and sites. The invention also includes a validation mechanism by crowd-sourcing and fake review generation to ensure reliability and trustworthiness of the scoring.
Abstract translation: 使用同一产品或服务（例如酒店，餐馆，诊所，发廊等）的多个评论来源来提供可信度得分。 这样的分数可以清楚地识别具有审查操作，遗漏和虚假的证据的酒店，并向用户全面了解产品或机构的审查。 在计算分数时使用三种类型的信息：空间，时间和网络或基于图形。 信息被混合以产生可靠地产生可信度得分的代表性特征集合。 本发明适应新的评论和网站。 本发明还包括通过采购和假审查生成的验证机制，以确保得分的可靠性和可信度。
-
公开(公告)号：US10089660B2
公开(公告)日：2018-10-02
申请号：US14849227
申请日：2015-09-09
Applicant: STC.UNM
Inventor： Shuang Luan , Abdullah Mueen , Michalis Faloutsos , Amanda J. Minnich
IPC: G06Q30/02
Abstract: Multiple sources of reviews for the same product or service (e.g. hotels, restaurants, clinics, hair saloon, etc.) are utilized to provide a trustworthiness score. Such a score can clearly identify hotels with evidence of review manipulation, omission and fakery and provide the user with a comprehensive understanding of the reviews of a product or establishment. Three types of information are used in computing the score: spatial, temporal and network or graph-based. The information is blended to produce a representative set of features that can reliably produce the trustworthiness score. The invention is self-adapting to new reviews and sites. The invention also includes a validation mechanism by crowd-sourcing and fake review generation to ensure reliability and trustworthiness of the scoring.
-
公开(公告)号：US10853372B1
公开(公告)日：2020-12-01
申请号：US15829575
申请日：2017-12-01
Applicant: STC.UNM
Inventor： Abdullah Mueen , Hossein Hamooni
IPC: G06F11/30 , G06F16/2455 , G06F16/27 , G06F40/103
Abstract: This disclosure describes technologies for pattern matching in data streams. Given one or more patterns of data, and one or more data streams, e.g., streams of measurements made by one or more sensors, this disclosure provides apparatus and techniques for identifying occurrences of the patterns in the data streams. Techniques according to this disclosure generally make use of distribution strategies, as disclosed herein, to distribute the work of pattern identification among multiple processors. Data streams may be divided into multiple segments, and patterns for identification as well as data stream segments may be distributed among the multiple processors.
-
公开(公告)号：US10389745B2
公开(公告)日：2019-08-20
申请号：US15750236
申请日：2016-08-04
Applicant: STC.UNM
Inventor： Abdullah Mueen , Nikan Chavoshi
Abstract: Bots are detected real-time by correlating activity between users using a lag-sensitive hashing technique that captures warping-invariant correlation. Correlated users groups in social media may be found that represent bot behavior with thousands of bot accounts detected in a couple of hours.
-
-
-
Search Results
4
records
(took 0.019 seconds)
