i need help with my research seminars and methodology

Anonymous
account_balance_wallet $10

Question Description

regarding the seminar on hate speech, you need to provide maximum one page (between half a page to one) summary of the talk (based on your notes during the talk and on the material I uploaded)

Deadline is Thursday April 5th.

Unformatted Attachment Preview

MANDOLA D3.1 Rights, Equality and Citizenship (REC) Programme of the European Commission (2014-2020) Monitoring and Detecting Online Hate Speech D3.1: MANDOLA Monitoring Dashboard† Abstract: The aim of this document is to present the implementation of MANDOLA monitoring dashboard, as well as to be used as a user manual. A detailed study of the architecture and the hate speech detection analysis is presented with a description on the implementation and usage of dashboard’s content. Each of the dashboard’s pages is described and the techniques used for better user experience are explained. Contractual Date of Delivery Actual Date of Delivery Deliverable Security Class Editors Contributors Quality Assurance † September 2016 September 2016 Public D. Stefanidis, D. Paschalides, G. Pallis All MANDOLA partners M. D. Dikaiakos This project is funded by the Rights, Equality and Citizenship (REC) Programme of the European Commission. www.MANDOLA-project.eu -1- September 30, 2016 MANDOLA D3.1 The MANDOLA consortium consists of: FORTH ACONITE ICITA INTHEMIS UAM UCY UM1 www.MANDOLA-project.eu Coordinator Principal Contractor Principal Contractor Principal Contractor Principal Contractor Principal Contractor Principal Contractor -2- Greece Ireland Bulgaria France Spain Cyprus France September 30, 2016 MANDOLA D3.1 Document Revisions & Quality Assurance Internal Reviewers Revisions Version 0.1 Date 10 September 2016 0.5 22 September 2016 1.0 29 September 2016 www.MANDOLA-project.eu By D. Paschalides, D. Stefanidis, M. Hernando D. Paschalides, D. Stefanidis, M. Hernando, G. Pallis, M. Dikaiakos D. Paschalides, D. Stefanidis, M. Hernando, G. Pallis, M. Dikaiakos -3- Overview MANDOLA Monitoring Dashboard MANDOLA Monitoring Dashboard MANDOLA Monitoring Dashboard September 30, 2016 MANDOLA D3.1 Table of Contents DOCUMENT REVISIONS & QUALITY ASSURANCE ............................................................... 3 TABLE OF CONTENTS ......................................................................................................... 4 TABLE OF FIGURES ............................................................................................................ 6 1 INTRODUCTION ......................................................................................................... 7 2 MONITORING DASHBOARD ARCHITECTURE ............................................................... 9 3 DATA STREAMS COLLECTION ................................................................................... 11 3.1 TWITTER DATA STREAM COLLECTION ...................................................................................... 11 3.2 GOOGLE DATA STREAM COLLECTION ...................................................................................... 12 3.3 DATA STREAM PROCESSING ................................................................................................. 15 4 DATA ANALYSIS ....................................................................................................... 16 4.1 MULTI-LINGUAL CORPUS .................................................................................................... 16 4.2 HATE-SPEECH DATA ANALYSIS .............................................................................................. 18 5 DATA STORAGE ....................................................................................................... 21 5.1 DATABASE........................................................................................................................ 21 5.2 DATA COLLECTIONS ........................................................................................................... 21 6 API .......................................................................................................................... 23 7 DASHBOARD ........................................................................................................... 24 7.1 HATE-MAP ....................................................................................................................... 24 7.1.1 Heat-map.............................................................................................................. 24 7.1.1.1 Geo-clustering .................................................................................................. 25 7.1.1.2 Date range bar .................................................................................................. 26 7.1.1.3 Filtering Categories .......................................................................................... 27 7.1.2 Hot-spot density map ........................................................................................... 28 7.1.2.1 Hate-rate metric ............................................................................................... 29 7.1.2.2 Date range filtering .......................................................................................... 29 7.1.2.3 Categories filtering ........................................................................................... 30 7.1.2.4 Country drill-down ........................................................................................... 31 7.2 STATISTICS ....................................................................................................................... 31 7.2.1 Time-line chart...................................................................................................... 32 7.2.1.1 Zoom functionality ........................................................................................... 32 7.2.1.2 Data zoom aggregation .................................................................................... 33 7.2.2 Language usage chart .......................................................................................... 34 7.2.3 Hate rate per category chart ................................................................................ 35 7.2.4 Hate rate per country chart .................................................................................. 35 7.2.5 Hate rate per city chart ........................................................................................ 36 7.2.6 Countries per category chart ................................................................................ 37 7.2.7 Cities per category chart ...................................................................................... 38 7.2.8 Time-line per category chart ................................................................................ 38 7.2.9 Hate strength gauge ............................................................................................ 39 7.3 RESPONSIVENESS AND MOBILE COMPATIBILITY ....................................................................... 39 www.MANDOLA-project.eu -4- September 30, 2016 MANDOLA D3.1 8 CONCLUSION ........................................................................................................... 40 REFERENCES ................................................................................................................... 41 ANNEX 1. API RESOURCES .............................................................................................. 42 www.MANDOLA-project.eu -5- September 30, 2016 MANDOLA D3.1 Table of Figures Figure 1: Mandola Dashboard Architectural Diagram ............................................................. 10 Figure 2: Ucy Framework Architecture .................................................................................... 12 Figure 3: General Activities Diagram - Modules Structures And Links Providers .................... 13 Figure 4: Activities Diagram - Services And Tools .................................................................... 14 Figure 5: Activities Diagram - Web Crawling Method .............................................................. 14 Figure 6: Kafka Message Queuing ............................................................................................ 15 Figure 7: Hate Filtering ............................................................................................................. 16 Figure 8: Heatmap Interface .................................................................................................... 24 Figure 9: Heatmap Gradient ..................................................................................................... 25 Figure 10: Geohash Definition Diagram ................................................................................... 25 Figure 11: Heatmap Zoom Out ................................................................................................. 26 Figure 12: Heatmap Zoom In .................................................................................................... 26 Figure 13: Date Range Bar Stable ............................................................................................. 26 Figure 14: Date Range Bar Moved ........................................................................................... 27 Figure 15: Hotspot Map Interface ............................................................................................ 28 Figure 16: Hotspot Map Gradient ............................................................................................ 28 Figure 17: Calendar Panel For Date Selection .......................................................................... 30 Figure 18: Statistics Date/Country Select Bar .......................................................................... 31 Figure 19: Timeline Hate-Rate Chart. Axis X Is Date And Axis Y Is Hate Rate (%) .................... 32 Figure 20: Timeline Zoom In Functionality............................................................................... 32 Figure 21: Timeline With Aggregated Results Per Date ........................................................... 34 Figure 22: Hate Rate Pie Chart ................................................................................................. 34 Figure 23: Disable Language Functionality............................................................................... 35 Figure 24: Hate Rate Per Category ........................................................................................... 35 Figure 25: Hate Rate Per Country ............................................................................................ 36 Figure 26: Disable Country Functionality ................................................................................. 36 Figure 27: Hate Categories Bubble Chart ................................................................................. 37 Figure 28: Top Three Countries Per Category .......................................................................... 37 Figure 29: Top Three Cities Per Category ................................................................................. 38 Figure 30: Timeline Per Category ............................................................................................. 38 Figure 31: Hate Strength Gauge ............................................................................................... 39 Figure 32: Mandola Dashboard Mobile Interface .................................................................... 39 Table 1: Hate Categories Table ................................................................................................ 18 Table 2: Hatespeech Data Collection Properties ..................................................................... 22 Table 3: Colour Representation In Heat Gradient.................................................................... 25 Table 4: Heat Map Filtering Category Buttons ......................................................................... 27 Table 5: Colour Representation In Heat Gradient.................................................................... 28 Table 6: Hotspot Map Category Filtering Buttons ................................................................... 30 Table 7: Data Aggregation Conditions...................................................................................... 33 Equation 1: Naïve Bayes Equation Equation 2: Hate Rate Metric Equation www.MANDOLA-project.eu 19 29 -6- September 30, 2016 MANDOLA D3.1 1 Introduction The term "hate speech" covers all forms of expression which spread, incite, promote, propagate, support or justify every form of hatred, violence, discrimination, segregation, hostility against persons or against a religion or the divine, including, inter alias, intolerance expressed by aggressive nationalism and ethnocentrism, discrimination and hostility against religious groups, minorities, migrants and people of immigrant origin, incitement or threat to commit harm, an offence or a crime, humiliation and offense to the dignity, insult, defamation and harassment. Although the definition lists a number of groups, which are frequently seen to be the targets of hate speech it does not limit the possible targets to these groups alone. This is an ‘openended’ definition (Deliverable 2.1 deals with proposing a definition for “hate speech”), in accordance with the open-ended understanding of discrimination adopted by the European Court of Human Rights. The relevant actions of hate speech should usually have been performed on the ground of the belonging or not belonging of the victim to a real or supposed particular group or on the ground of one of the personal characteristics of the victim, which might be physical (e.g. colour, handicap), psychological, philosophical (e.g. religion, beliefs), or behavioural (e.g. exercise of a worship, belonging to a professional organisation) or have been committed against religion or the divine. In recent years an ominous picture about online hate speech has started to materialise within cyberspace. Recent polls suggest that as many as four out of five respondents have encountered hate speech online and two out of five have personally felt attacked or threatened [1]. Although it is difficult to get accurate statistics about the spread of hate speech in cyberspace, the picture is becoming increasingly clear: the Internet is alarmingly effective at spreading hate speech – so much so that most Internet users have encountered it at some point. To make matters worse, hate speech usually targets the most vulnerable groups within society: children, minorities, and immigrants – groups that by definition have little capacity to protect themselves, both in the online and the physical worlds. There are two major difficulties in dealing with online hate speech: (i) Lack of reliable data that can show detailed online hate speech trends. (ii) Poor awareness about how to deal with the issue since there is a fine line between hate speech and freedom of speech: the boundaries between “legal” and potentially illegal hate speech are sometimes blurred, and may vary between territories. The same speech might not be illegal in all E.U. countries, and might in some countries be illegal only if some additional circumstances are noticed. MANDOLA plans to fill this gap by: (i) monitoring the spread and penetration of online hate-related speech in Europe and in member states using big data approaches; (ii) providing policy makers with actionable information that can be used to promote policies that mitigate the spread of online hate speech; (iii) providing ordinary citizens with useful tools that can help them deal with online hate speech; (iv) transferring best practices among member states. www.MANDOLA-project.eu -7- September 30, 2016 MANDOLA D3.1 In this deliverable, the multi-lingual monitoring dashboard1 is presented, which has been developed, in order to offer reliable information about online hate speech enabling users to focus on their geographic region ranging from their city to their country to the entire European Union. The dashboard uses Twitter and Web sites as sources of possible hate-related online content. The rest of this deliverable is structured as follows: Section 2 presents the architecture of the monitoring system. The data stream collection mechanism is presented in Section 3. Section 4 describes the data processing procedure that takes place. Section 5 presents how data is stored. Section 6 presents the main features of API. Section 7 presents the MANDOLA Monitoring dashboard functionalities and Section 8 concludes this deliverable. 1 Currently available on http://mandola.grid.ucy.ac.cy:3000 www.MANDOLA-project.eu -8- September 30, 2016 MANDOLA D3.1 2 Monitoring Dashboard Architecture The monitoring dashboard in MANDOLA handles two types of data streams collections. The first one is the Twitter data stream and the second one is the Google data stream. The data streams are handled through a distributed publish-subscribe messaging system named Apache Kafka [2]. Kafka feeds the hate-speech data analysis module and also collects a data sample set in order to create the multi-lingual corpus based on hate filtering module. The hate-speech data analysis module utilizes sentiment analysis tools via the NLTK platform [3] in order to classify content whether it is hate-related speech or not. The multi-lingual corpus, which is used to train the classification model that exists in the hate-speech data analysis module, is given by experts, which are called social scientists. Social scientists classify the hate-related content in the given Twitter and Google sample set, based on its strength and its categories. The hate filtering module is used to conduct automatically an initial filtering of the sample set so that social scientists receive for review more relevant content (i.e., hate-related). A multi-lingual corpus for hate speech is given as input to the hate filtering module. The corpus has been initially built from hate databases such as the crowdsourcing database Hatebase [4], and from hate-related sentiment lexicons containing seed words such as the AFINN [5] lexicon. Specifically, Hatebase is a Canadian initiative that its ultimate goal is to build the largest online repository of structured, multilingual, usage-based hate speech. Hatebase consists of hate words and phrases, annotated with their offensiveness strength, their category and their meaning. The multi-lingual corpus is enriched by the input of social scientists. The corpus is given as input to hate-speech data analysis module. Section 4 describes the data processing procedure that takes place. When the processing is done, the output is stored in the hate speech database (MongoDB). More details about the data storage are described in section 5. The dashboard is connected with the hate speech database (MongoDB) via an API that is used to retrieve data required for the various types of data visualization supported by the Dashboard (heat map, charts etc.). The API functionalities and resources are presented in section 6. Figure 1 depicts the architecture of the MANDOLA Monitoring Dashboard. For the implementation of the dashboard, the Express application framework [6] has been used. Express is a minimal and flexible Node JS application framework that provides a robust set of features for web and mobile development. Express supports a thin layer of fundamental web application features, without obscuring Node JS features, with HTTP utility methods and middleware to support API implementation. www.MANDOLA-project.eu -9- September 30, 2016 MANDOLA D3.1 Figure 1: MANDOLA Dashboard Architectural Diagram www.MANDOLA-project.eu - 10 - September 30, 2016 MANDOLA D3.1 3 Data streams collection The Data collection engine consists of two sub-modules that are responsible for collecting data from Twitter [7] and Google API [8]. This is done for purely research purposes in the MANDOLA project. The processing and storing is in line with article 7(1)(2) of the Personal Data Protection (Protection ...
Purchase answer to see full attachment

Tutor Answer

Robert__F
School: Duke University

Good luck in your study and if you need any further help in your assignments, please let me know Can you please confirm if you have received the work? Once again, thanks for allowing me to help you R MESSAGE TO STUDYPOOL NO OUTLINE IS NEEDED AS IT IS A SPEECH

Running head: ENGLISH HOMEWORK

Hate Speech
Student’s name
University affiliation

ENGLISH HOMEWORK

2

Hate speech has been termed as anything that covers all forms of expression
which spread, incites, justifies...

flag Report DMCA
Review

Anonymous
awesome work thanks

Brown University





1271 Tutors

California Institute of Technology




2131 Tutors

Carnegie Mellon University




982 Tutors

Columbia University





1256 Tutors

Dartmouth University





2113 Tutors

Emory University





2279 Tutors

Harvard University





599 Tutors

Massachusetts Institute of Technology



2319 Tutors

New York University





1645 Tutors

Notre Dam University





1911 Tutors

Oklahoma University





2122 Tutors

Pennsylvania State University





932 Tutors

Princeton University





1211 Tutors

Stanford University





983 Tutors

University of California





1282 Tutors

Oxford University





123 Tutors

Yale University





2325 Tutors