DOCSLIB.ORG

UCLA Electronic Theses and Dissertations

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
UCLA Electronic Theses and Dissertations UCLA UCLA Electronic Theses and Dissertations Title Probabilistic Topic Models for Graph Mining Permalink https://escholarship.org/uc/item/7ss082g1 Author Cha, Young Chul Publication Date 2014 Peer reviewed|Thesis/dissertation eScholarship.org Powered by the California Digital Library University of California UNIVERSITY OF CALIFORNIA Los Angeles Probabilistic Topic Models for Graph Mining A dissertation submitted in partial satisfaction of the requirements for the degree Doctor of Philosophy in Computer Science by Young Chul Cha 2014 c Copyright by Young Chul Cha 2014 ABSTRACT OF THE DISSERTATION Probabilistic Topic Models for Graph Mining by Young Chul Cha Doctor of Philosophy in Computer Science University of California, Los Angeles, 2014 Professor Junghoo Cho, Chair In this research, we extend probabilistic topic models, originally developed for a tex- tual corpus analysis, to analyze a more general graph. Especially, we extend them to effectively handle: (1) a bias caused by a limited number of frequent nodes (“popular- ity bias”), and (2) complex graphs having more than two entity types. For the popularity bias problem, we propose LDA extensions and new topic mod- els explicitly modeling the popularity of a node with a “popularity component”. In extensive experiments with a real-world Twitter dataset, our approaches achieve signif- icantly lower perplexity (i.e., better prediction power) and improved human-perceived clustering quality compared to LDA. To analyze more complex graphs, we propose a novel universal topic framework that takes an “incremental” approach of breaking a complex graph into smaller units, learning the topic group of each entity from the smaller units, and then “propagating” the learned topics to others. In a DBLP prediction problem, our approach achieves the best performance over many state-of-the-art methods. We also demonstrate huge potential of our approach with search logs from a commercial search engine. ii The dissertation of Young Chul Cha is approved. Carlo Zaniolo D. Stott Parker Gregory H. Leazer Junghoo Cho, Committee Chair University of California, Los Angeles 2014 iii To my family . iv TABLE OF CONTENTS 1 Introduction ::::::::::::::::::::::::::::::::: 1 1.1 Challenges in Graph Mining Using Probabilistic Topic Models . 1 1.2 Organization of Dissertation . 4 2 Preliminaries :::::::::::::::::::::::::::::::: 6 2.1 Probabilistic Topic Models . 6 2.2 Heterogeneous Information Networks . 8 3 Handling a Popularity Bias in Topic Models ::::::::::::::: 10 3.1 Introduction . 10 3.2 Applying Topic Models to Social Graphs . 13 3.2.1 Follow-Edge Generative Model . 13 3.2.2 Popularity Bias . 17 3.3 Handling a Popularity Bias . 19 3.3.1 Existing LDA Extensions . 21 3.3.2 Procedural Variations of LDA . 23 3.3.3 Popularity-Aware Topic Models . 26 3.4 Experiments . 31 3.4.1 Dataset and Experimental Settings . 32 3.4.2 Prediction Performance Analysis . 35 3.4.3 Example Topic Groups . 37 3.4.4 Grouping Quality Analysis . 40 v 3.5 Conclusion . 43 4 Complex-Graph Analysis Using Topic Models :::::::::::::: 44 4.1 Introduction . 44 4.2 Universal Topic Framework . 48 4.2.1 Edge Generative Model . 48 4.2.2 Incorporating Learned Topics . 51 4.2.3 Issues in Topic Incorporation . 54 4.3 Experiments and Analyses . 56 4.3.1 DBLP Experiment . 57 4.3.2 Online Search Experiments . 62 4.4 Conclusion . 70 5 Related Work :::::::::::::::::::::::::::::::: 73 6 Conclusions and Future Work ::::::::::::::::::::::: 76 References ::::::::::::::::::::::::::::::::::: 78 vi LIST OF FIGURES 3.1 Bipartite graph representation of models . 14 3.2 Topic group on cycling showing the popularity bias . 18 3.3 Topic hierarchy and documents generated in the hierarchy . 22 3.4 Two-step labeling approach . 24 3.5 Example of the threshold noise filtering process . 25 3.6 LDA and proposed topic models . 27 3.7 Distributions of incoming and outgoing edges . 33 3.8 Perplexity comparison . 36 3.9 Sample topic groups I . 38 3.10 Sample topic groups II . 39 3.11 Quality comparison . 42 4.1 Examples of HINs . 46 4.2 Example of edge labeling . 50 4.3 Proposed topic models . 51 4.4 Structures of two types of search logs . 62 4.5 Examples of topic incorporation orders . 64 4.6 Performance Analysis . 65 4.7 Topic Granularity Analysis . 69 4.8 Incorporation of C-Log (the last bars show results using C-Log) . 71 vii LIST OF TABLES 3.1 Symbols used throughout this chapter and their meanings . 20 3.2 Statistics of our Twitter dataset . 33 3.3 Experimental cases and descriptions . 35 4.1 Statistics of the datasets . 58 4.2 Prediction accuracy on the DBLP dataset. Except ours, all other results are from [16, 12]. 60 4.3 Venue clusters . 61 4.4 Example topic cluster without UniZ (topic number 43) . 67 4.5 Example topic cluster from UniZ . 68 viii ACKNOWLEDGMENTS First, I would like to thank my wonderful advisor, Junghoo “John” Cho. If it had not been for his help and advice, I would not have finished my Ph.D. study. When I came back to school after working for 10 years in a telecom industry in 2009, I had a quite hard time to catch up with all the coursework required for my new major, data mining. He was very patient with my slow progress and has never rushed me to do something. Whenever I was struggling with a hard problem, he provided me with an insightful advice rather than a solution, and motivated me to learn while seeking for the solution. I remember his I-know-it-all smile when I made a great fuss about my small finding after following his advice. Although I was very busy with my school work during my first two years, I was happy because I could feel I was getting better and better as a researcher. I really appreciate his help and great advice. I only hope that I will be able to practice half as much as what I learned from him when I begin to work with mentees of my own. Second, I thank my lovely wife, Gook Hee Lee, for her great support. I know that it was not an easy decision for her to quit her prospective job in Korea and live as a wife of a graduate student in the foreign country. However, she went through all the hard years without many complaints. She was very supportive and willingly adjusted her schedule to mine. I especially thank her for her effort to make me healthier when I had a health issue. She searched good food ingredients for me and spent time to learn how to make delicious food with them. I also thank her for her great culinary skills. Without them, I would have had hard times to eat a large amount of food she cooks for me. Special thanks to Carlo Zaniolo, Stott Parker and Gregory Leazer for serving as my committee. Their insightful comments made my thesis more solid and I learned a lot from their comments. Especially, Greg gave me very detailed comments on various ix aspects of my thesis. I also owe Carlo, Stott and John for their great classes. I could establish good data mining knowledge with their wonderful classes and projects. I would also like to thank our group members, including Michael Welch, Uri Schonfeld, Chu-Cheng Hsieh, Bin Bi, Yuchen Liu, Dong Wang, Zijun Xue, Christo- pher Moghbel, and Giljoo Na. Michael and Uri gave me good advices when I first joined the group, and Chu-Cheng always magically solved my computer and network issues. Bin was so smart and helped me a lot with developing topic models, and I had many interesting academic talks with Yuchen. Dong and Zijun were always nice to me whenever I need their help. Chris proofread most of my papers and provided sharp comments. Giljoo gave me good life advices and his wife frequently invited me to wonderful dinners when I lived alone during my first year. It was very lucky for me to have chances to work with Keng-hao Chang, Hari Bommaganti, Ye Chen, Jian Yuan and Tak Yan during my internships at Microsoft. They provided great ideas and a stimulating environment during the summers. I could also practice what I learned from school with large real datasets. Finally, I would like to thank my parents and brothers for their invariable love and support. They always firmly believe me and respect my decision. Whenever I am weary and in doubt, their love and belief cheer me up and give me strength. I am grateful for their selfless support and hope they are always healthy and happy. I deeply thank God who takes care of my family and leads me to the right way. x VITA 1998 B.S. (Computer Science), Yonsei University, Korea. 2000 M.S. (Computer Science), Yonsei University, Korea. 2000–2014 Researcher & Manager, KT, Korea. PUBLICATIONS Y. Cha, B. Bi, and J. Cho, Handling a Popularity Bias in Probabilistic Topic Models, under review Y. Cha, B. Bi, J. Cho, K. Chang, H. Bommaganti, Y. Chen, and T. Yan, A Universal Topic Framework (UniZ) and Its Application in Online Search, under review Y. Cha, J. Cho, J. Yuan, and T. Yan, Exploration of the Effects of Category Match Score in Search Advertising, ICDE, Apr. 2014. Y. Cha, B. Bi, C. Hsieh, and J. Cho, Incorporating Popularity in Topic Models for Social Network Analysis, SIGIR, best paper award nominee, Jul.-Aug. 2013. Y. Cha, and J. Cho, Social Network Analysis Using Topic Models, SIGIR, Aug. 2012. xi CHAPTER 1 Introduction 1.1 Challenges in Graph Mining Using Probabilistic Topic Models Recently, topic models have widely been used for a textual corpus analysis due to their high-quality analysis results.
Load more

Recommended publications
  • Causal Consistency and Latency Optimality: Friend Or Foe?
    Causal Consistency and Latency Optimality: Friend or Foe? Diego Didona1, Rachid Guerraoui1, Jingjing Wang1, Willy Zwaenepoel1;2 1EPFL, 2 University of Sydney first.last@epfl.ch ABSTRACT 1. INTRODUCTION Causal consistency is an attractive consistency model for Geo-replication is gaining momentum in industry [9, 16, geo-replicated data stores. It is provably the strongest model 20, 22, 25, 44, 51, 52, 66] and academia [24, 35, 48, 50, 60, that tolerates network partitions. It avoids the long laten- 70, 71, 72] as a design choice for large-scale data platforms cies associated with strong consistency, and, especially when to meet the strict latency and availability requirements of using read-only transactions (ROTs), it prevents many of on-line applications [5, 56, 63]. the anomalies of weaker consistency models. Recent work Causal consistency. To build geo-replicated data stores, has shown that causal consistency allows \latency-optimal" causal consistency (CC) [2] is an attractive consistency model. ROTs, that are nonblocking, single-round and single-version On the one hand, CC has an intuitive semantics and avoids in terms of communication. On the surface, this latency op- many anomalies that are allowed under weaker consistency timality is very appealing, as the vast majority of applica- models [25, 68]. On the other hand, CC avoids the long tions are assumed to have read-dominated workloads. latencies incurred by strong consistency [22, 32] and toler- In this paper, we show that such \latency-optimal" ROTs ates network partitions [41]. CC is provably the strongest induce an extra overhead on writes that is so high that consistency level that can be achieved in an always-available it actually jeopardizes performance even in read-dominated system [7, 45].
    [Show full text]
  • DME Server Administration Reference
    DME Server Administration Reference DME version 4.2SP2 Document version 1.1 Published on 17. September 2014 DME Server Administration Reference Introduction Contents Introduction ............................................................................ 11 Copyright information ......................................................................... 11 Company information .......................................................................... 12 Typographical conventions.................................................................. 12 About DME ........................................................................................... 13 Features and benefits .............................................................................. 13 About the DME Server ........................................................................ 14 Supported platforms ................................................................................ 15 DME server architecture ........................................................ 17 One server, many connectors ............................................................ 17 The DME server ................................................................................... 19 The connector ...................................................................................... 20 The group graph ....................................................................................... 21 Load balancing and failover .................................................................... 22 The AppBox
    [Show full text]
  • Causal Consistency and Latency Optimality: Friend Or Foe?
    Causal Consistency and Latency Optimality: Friend or Foe? Diego Didona, Rachid Guerraoui, Jingjing Wang, Willy Zwaenepoel EPFL first.last@epfl.ch ABSTRACT the long latencies incurred by strong consistency (e.g., lin- Causal consistency is an attractive consistency model for earizability and strict serializability) [30, 21] and tolerates replicated data stores. It is provably the strongest model network partitions [37]. In fact, CC is provably the strongest that tolerates partitions, it avoids the long latencies as- consistency that can be achieved in an always-available sys- sociated with strong consistency, and, especially when us- tem [41, 7]. CC is the target consistency level of many ing read-only transactions, it prevents many of the anoma- systems [37, 38, 25, 26, 4, 18, 29], it is used in platforms lies of weaker consistency models. Recent work has shown that support multiple levels of consistency [13, 36], and it that causal consistency allows \latency-optimal" read-only is a building block for strong consistency systems [12] and transactions, that are nonblocking, single-version and single- formal checkers of distributed protocols [28]. round in terms of communication. On the surface, this la- Read-only transactions in CC. High-level operations tency optimality is very appealing, as the vast majority of such as producing a web page often translate to multiple applications are assumed to have read-dominated workloads. reads from the underlying data store [47]. Ensuring that all In this paper, we show that such \latency-optimal" read- these reads are served from the same consistent snapshot only transactions induce an extra overhead on writes; the avoids undesirable anomalies, extra overhead is so high that performance is actually jeop- in particular, the following well-known anomaly: Alice re- ardized, even in read-dominated workloads.
    [Show full text]
  • RSS Readers & Customized Dashboards
    Legal Services National Technology Assistance Project www.lsntap.org RSS Readers & Customized Dashboards RSS stands for "Rich Site Summary" and it is a dialect of XML. Although the technical definition of RSS isn't the easiest to understand, don't let that scare you away from this useful timesaving tool. If you are unfamiliar with RSS, I recommend this short video: RSS in Plain English. It explains RSS simply and shows you how to subscribe to feeds by looking for this logo on your favorite blogs and websites. In addition to the orange RSS icon, look for these icons as well Once you've subscribed to several sites, you view them in your RSS reader of choice. The video recommends using Google Reader which no longer exists. Below is a list of RSS readers and dashboards to help you keep all of your RSS feeds in one convenient location. Most RSS readers allow you to subscribe to feeds directly from their website by simply typing in a URL, ie www.lsntap.org will give you the headlines for all of LSNTAP's blog posts. A lot of RSS readers have additional features which allow you to create customized dashboards with RSS feeds, email, weather, etc. There are many more RSS readers out there. If one of these isn't exactly what you are looking for, I suggest doing some more research. Here is a list of RSS readers that you might find useful. Netvibes/Bloglines (Great Free version) Both sites allows you to create multiple dashboards. You can have a dashboard for work, home, or for different projects.
    [Show full text]
  • DME Client User Guide 4.1.5 for Apple Ios Devices
    DME Client User Guide 4.1.5 For Apple iOS devices Document version 1.0 Created on 17-09-2013 Contents Introduction .............................................................................. 4 Copyright information ........................................................................... 4 Company information ............................................................................ 5 About DME.............................................................................................. 6 Features and benefits ................................................................................ 6 Terminology ............................................................................................ 7 General concepts ...................................................................... 9 Data security ........................................................................................... 9 Changing mailbox passwords ................................................................. 10 Switching users ......................................................................................... 11 Synchronization overview ................................................................... 12 What is synchronized .............................................................................. 12 Conflicts ..................................................................................................... 13 Synchronization methods ........................................................................ 13 Synchronization manager ......................................................................
    [Show full text]
  • Firefox Rss Feed Notification
    Firefox Rss Feed Notification Nicky reprices his sociopaths plod briskly, but snootiest Krishna never abridged so round-arm. Which Chandler headquarter so outwards that Magnum demark her acronym? Ganglier and etymological Rufus tastes while explicable Quill fool her repleteness actinically and tellurized flop. Feeder Get this Extension for Firefox en-US. Does Firefox support RSS? Provide an RSS feed be an alternative to the email notifications. RSS Feeds Overview Powered by Kayako fusion Help Desk. You are rss feeds from firefox, they probably the entire ui. RSS reader setup examples Intel. The feed for basic features than visiting each member yet? It does not be able to use the technology we want to divide feeds the old messages with the browser forks where he want. Download Feedly Notifier for Firefox A lightweight yet not useful Firefox extension that keeps you west to spread with the RSS feed while your. Feedbro Get this Extension for Firefox en-US. RSS Really Simple Syndication feeds provide news headlines brief article. Simple RSS notifier mozillaZine Forums. List manually check this rss feeds, firefox is for professionals who published, firefox rss anymore then automatically enrolled in madison, you need a feature is. I could also copper the RSS to notifyme and get SMS notifications or updates directly to. 5 Best RSS feed reader extensions or applications that. You choose the. Of notifications of the notification support in my hands, it wakes up in ubuntu software centre of what is the feed? Choose how to answer your web push notifications title first and notification text. If you want frequent use the copy and past url method choose preview in Firefox This will.
    [Show full text]
  • Gérez Vos Flux Librement Grâce À Kriss Et Leed
    Logiciel libre Gérez vos flux librement grâce à KrISS et Leed Raphael.Grolimund@epfl.ch, EPFL, bibliothécaire & [email protected], HEG Genève, filière Information documentaire, assistant d’enseignement en informatique documentaire tions d’une partie des utilisateurs ont révélé, ou du moins rappelé: Google Reader closed, but it doesn’t mean that RSS z qu’il y a un public, peut-être minoritaire, mais significatif, qui is dead. This article presents two free online self- se sert de cette technologie; hosted solutions to get rid of commercial third-party z que Google Reader répondait efficacement à une demande; dependency. z que la dépendance à un service Web proposé par un tiers peut poser problème. Ce n’est pas parce que Google Reader a fermé que le Pour comprendre ces trois points, il n’est pas inutile de rappeler RSS est mort. Cet article présente deux logiciels en le fonctionnement des flux RSS et les différents outils qui per- ligne libres à héberger pour sortir de la dépendance mettent de s’en servir. vis-à-vis d’un prestataire commercial. Qu’est-ce que le RSS ? Fiche descriptive Le RSS est une technologie qui dispense l’utilisateur de visiter un site Web pour savoir s’il y a des nouveautés. L’information vient KrISS & Leed à l’utilisateur via la mise à jour du flux RSS. Grâce aux flux RSS, il Domaine est donc possible et assez facile de suivre l’actualité de plusieurs ✦ Lecture et gestion de flux RSS dizaines, voire centaines, de sites Web. L’acronyme RSS a tour à tour signifié RDF Site Summary, Rich Site Licence KrISS langue KrISS version KrISS Summary et Really Simple Syndication.
    [Show full text]
  • OSINT Handbook September 2020
    OPEN SOURCE INTELLIGENCE TOOLS AND RESOURCES HANDBOOK 2020 OPEN SOURCE INTELLIGENCE TOOLS AND RESOURCES HANDBOOK 2020 Aleksandra Bielska Noa Rebecca Kurz, Yves Baumgartner, Vytenis Benetis 2 Foreword I am delighted to share with you the 2020 edition of the OSINT Tools and Resources Handbook. Once again, the Handbook has been revised and updated to reflect the evolution of this discipline, and the many strategic, operational and technical challenges OSINT practitioners have to grapple with. Given the speed of change on the web, some might question the wisdom of pulling together such a resource. What’s wrong with the Top 10 tools, or the Top 100? There are only so many resources one can bookmark after all. Such arguments are not without merit. My fear, however, is that they are also shortsighted. I offer four reasons why. To begin, a shortlist betrays the widening spectrum of OSINT practice. Whereas OSINT was once the preserve of analysts working in national security, it now embraces a growing class of professionals in fields as diverse as journalism, cybersecurity, investment research, crisis management and human rights. A limited toolkit can never satisfy all of these constituencies. Second, a good OSINT practitioner is someone who is comfortable working with different tools, sources and collection strategies. The temptation toward narrow specialisation in OSINT is one that has to be resisted. Why? Because no research task is ever as tidy as the customer’s requirements are likely to suggest. Third, is the inevitable realisation that good tool awareness is equivalent to good source awareness. Indeed, the right tool can determine whether you harvest the right information.
    [Show full text]
  • Improved User News Feed Customization for an Open Source Search Engine
    San Jose State University SJSU ScholarWorks Master's Projects Master's Theses and Graduate Research Spring 5-8-2020 Improved User News Feed Customization for an Open Source Search Engine Timothy Chow San Jose State University Follow this and additional works at: https://scholarworks.sjsu.edu/etd_projects Part of the Databases and Information Systems Commons Recommended Citation Chow, Timothy, "Improved User News Feed Customization for an Open Source Search Engine" (2020). Master's Projects. 908. DOI: https://doi.org/10.31979/etd.uw47-hp9s https://scholarworks.sjsu.edu/etd_projects/908 This Master's Project is brought to you for free and open access by the Master's Theses and Graduate Research at SJSU ScholarWorks. It has been accepted for inclusion in Master's Projects by an authorized administrator of SJSU ScholarWorks. For more information, please contact [email protected]. Improved User News Feed Customization for an Open Source Search Engine A Project Presented to The Faculty of the Department of Computer Science San José State University In Partial Fulfillment of the Requirements for the Degree Master of Science By Timothy Chow May 2020 1 © 2020 Timothy Chow ALL RIGHTS RESERVED 2 The Designated Project Committee Approves the Master’s Project Titled IMPROVED USER NEWS FEED CUSTOMIZATION FOR AN OPEN SOURCE SEARCH ENGINE by Timothy Chow APPROVED FOR THE DEPARTMENT OF COMPUTER SCIENCE SAN JOSÉ STATE UNIVERSITY May 2020 Dr. Chris Pollett Department of Computer Science Dr. Robert Chun Department of Computer Science Dr. Thomas Austin Department of Computer Science 3 ABSTRACT Yioop is an open source search engine project hosted on the site of the same name.It offers several features outside of searching, with one such feature being a news feed.
    [Show full text]
  • Favorite Culinary RSS Feeds
    Favorite Culinary RSS Feeds What is RSS and an RSS feed? RSS (Real Simple Syndication or Rich Site Summary) is a format for delivering regularly changing web content automatically. Many news-related sites, weblogs and other online publishers deliver their content as an RSS Feed to as it becomes available. Quick, automated information delivery enables the recipient to be more knowledgeable and competitive due to possessing what they need in a timely manner. Consequently, you are often requested to sign up for a desktop “Reader” so that the feed information can be delivered directly to your own computer in an organized fashion. In addition, choosing an appropriate feed search phrase can not only be useful but also lead to life-saving discoveries. See the Feed Phrase Examples below for ideas. Main Menu Feed Readers or Feed Search Engines (Online) Culinary Chowhound RSS Feeds Cooking Light RSS Feeds General RSS Micro - Real-Time Search Powered by FeedRank Ukora – News Search Feed Readers and Aggregators (Desktop – residing on your computer) Requiring Account Creation and / or Sign-in Best Free RSS Reader / Aggregator chefsteflux.com 1 facebook.com/chefsteflux Feed Readers and Aggregators (Continued) Comma Feed Feedly Feed Reader G2Reader Inoreader News360 Newstab RSS Owl The Old Reader Ten Best Readers for RSS, News and More Sample Feed Sites Cooking Light Eating Well My.Recipes Feeds Miscellaneous Feed Sites RSS Feed Phrase Examples Check the links below. Then, try your own feed search. Cooking Light Boost immunity Cancer prevention Digestive
    [Show full text]
  • Open Source Intelligence Tools and Resources Handbook.Pdf
    Open Source Intelligence Tools and Resources Handbook November 2016 Table of Contents Table of Contents ...................................................................................................................................... 2 General Search ........................................................................................................................................... 5 Main National Search Engines ................................................................................................................. 7 Meta Search ............................................................................................................................................... 8 Specialty Search Engines .......................................................................................................................... 9 Visual Search and Clustering Search Engines ..................................................................................... 10 Similar Sites Search ................................................................................................................................. 11 Document and Slides Search ................................................................................................................. 12 Pastebins ................................................................................................................................................... 13 Code Search ............................................................................................................................................
    [Show full text]
  • Chapter 5. Remote Monitoring
    Chapter 5: Remote Monitoring Chapter 5. Remote Monitoring A. Collecting Information from Diaspora Communities Diaspora communities do not simply shut the door to their countries upon resettling in a third country. Families remain behind, and emotional, economic, and political ties are strong. Many living in exile hope one day to return home. Consequently, members of diaspora communities can often be a good source for credible and current information about human rights country conditions. Many members of diaspora communities have first-hand knowledge of human rights abuses in their country of origin. Personal accounts of refugees and asylum seekers, as well as other immigrants, provide a window into the human rights conditions that forced them to leave their country. These accounts not only establish the basis for protection of individuals but also provide evidence to hold governments accountable for violations of international human rights obligations. Human Rights Fact-Finding with the Oromo Diaspora The Advocates for Human Rights used remote monitoring to research the 2009 report Human Rights in Ethiopia: Through the Eyes of the Oromo Diaspora. The Oromo are Ethiopia’s largest ethnic group, but have experienced repression and human rights abuses by successive Ethiopian governments since the late 19th century. With this report, The Advocates sought to give voice to the experiences of the Oromo people in the diaspora, to highlight a history of systematic political repression in Ethiopia, and to support the improvement of human rights conditions in Ethiopia. 67 Chapter 5: Remote Monitoring The project methodology focused on interviews with members of the Oromo diaspora, as well as service providers, scholars and others working with Oromos.
    [Show full text]