Economist Year in Review: Part 1

Intro: A new a year means a new Year in Review to digest and synthesize the events. Here's my attempt creating a Data oriented Year in Review of The Economist Newspaper using all the articles from the past year.

Getting Data: I used Python to 'scrape' data from The Economist website. I've included the code, but not the actual articles. If you have access to the Economist then you too can use the code to download the articles from the past year.

Scrape Code

LDA: First I wanted to know what kind of topics were written about so I made a topics model with 100 topics. Below are the top ten words of each topic in decreasing order of importance. In addition, some of topics displayed over time that I thought were interesting.

library(topicmodels)
load("/users/sweiss/google drive/ld100.Rda")
theta = posterior(ld.100)$topics
theta.average = colMeans(theta)
top.5.factors = names(sort(theta.average, decreasing = TRUE))
terms(ld.100, 10)[, as.numeric(top.5.factors)]

##       Topic 50    Topic 16   Topic 62    Topic 90  Topic 83    
##  [1,] "price"     "elect"    "scienc"    "bank"    "minist"    
##  [2,] "rate"      "parti"    "cell"      "financi" "govern"    
##  [3,] "economist" "vote"     "research"  "loan"    "polit"     
##  [4,] "index"     "polit"    "found"     "capit"   "prime"     
##  [5,] "market"    "voter"    "technolog" "lend"    "parti"     
##  [6,] "econom"    "poll"     "one"       "credit"  "parliament"
##  [7,] "exchang"   "seat"     "human"     "deposit" "opposit"   
##  [8,] "commod"    "candid"   "brain"     "financ"  "leader"    
##  [9,] "interest"  "win"      "look"      "asset"   "elect"     
## [10,] "job"       "campaign" "univers"   "regul"   "coalit"    
##       Topic 46     Topic 64  Topic 52    Topic 66     Topic 30    
##  [1,] "market"     "number"  "billion"   "presid"     "republican"
##  [2,] "firm"       "year"    "compani"   "polit"      "obama"     
##  [3,] "product"    "sinc"    "firm"      "elect"      "senat"     
##  [4,] "industri"   "increas" "share"     "power"      "democrat"  
##  [5,] "sale"       "rise"    "busi"      "presidenti" "congress"  
##  [6,] "manufactur" "rate"    "profit"    "year"       "hous"      
##  [7,] "make"       "fall"    "deal"      "govern"     "state"     
##  [8,] "profit"     "accord"  "buy"       "constitut"  "polit"     
##  [9,] "busi"       "quarter" "sharehold" "last"       "barack"    
## [10,] "cost"       "averag"  "stake"     "countri"    "presid"    
##       Topic 20  Topic 24  Topic 54   Topic 67 Topic 27  Topic 74 
##  [1,] "economi" "rate"    "group"    "syria"  "will"    "use"    
##  [2,] "growth"  "inflat"  "govern"   "rebel"  "may"     "system" 
##  [3,] "econom"  "bank"    "armi"     "war"    "year"    "can"    
##  [4,] "gdp"     "polici"  "peac"     "assad"  "hope"    "work"   
##  [5,] "invest"  "central" "pakistan" "regim"  "alreadi" "one"    
##  [6,] "year"    "currenc" "kill"     "weapon" "get"     "make"   
##  [7,] "account" "market"  "polit"    "iraq"   "next"    "way"    
##  [8,] "quarter" "fed"     "violenc"  "syrian" "expect"  "machin" 
##  [9,] "export"  "economi" "war"      "libya"  "think"   "design" 
## [10,] "spend"   "reserv"  "attack"   "forc"   "far"     "problem"
##       Topic 49  Topic 22     Topic 96   Topic 48   Topic 78   Topic 25 
##  [1,] "citi"    "parti"      "fund"     "firm"     "one"      "polic"  
##  [2,] "local"   "polit"      "investor" "compani"  "may"      "crime"  
##  [3,] "mayor"   "leader"     "return"   "busi"     "might"    "prison" 
##  [4,] "town"    "power"      "invest"   "consult"  "like"     "drug"   
##  [5,] "area"    "member"     "asset"    "industri" "way"      "say"    
##  [6,] "council" "politician" "equiti"   "big"      "can"      "crimin" 
##  [7,] "resid"   "call"       "manag"    "servic"   "whether"  "jail"   
##  [8,] "centr"   "nation"     "share"    "client"   "even"     "sentenc"
##  [9,] "street"  "offici"     "stock"    "work"     "question" "murder" 
## [10,] "build"   "chief"      "money"    "say"      "littl"    "gun"    
##       Topic 31   Topic 21   Topic 26   Topic 77    Topic 61 Topic 40
##  [1,] "euro"     "investig" "labour"   "mobil"     "year"   "year"  
##  [2,] "zone"     "charg"    "britain"  "technolog" "day"    "last"  
##  [3,] "spain"    "alleg"    "cameron"  "phone"     "week"   "two"   
##  [4,] "european" "claim"    "tori"     "appl"      "april"  "ago"   
##  [5,] "bail"     "case"     "parti"    "comput"    "month"  "five"  
##  [6,] "europ"    "trial"    "conserv"  "servic"    "last"   "past"  
##  [7,] "greec"    "report"   "miliband" "oper"      "may"    "month" 
##  [8,] "countri"  "scandal"  "mps"      "googl"     "time"   "four"  
##  [9,] "ireland"  "accus"    "david"    "network"   "said"   "three" 
## [10,] "cyprus"   "former"   "polit"    "softwar"   "june"   "end"   
##       Topic 33 Topic 5   Topic 11 Topic 23  Topic 15   Topic 80   Topic 29
##  [1,] "can"    "cut"     "court"  "money"   "britain"  "china"    "long"  
##  [2,] "make"   "spend"   "law"    "cost"    "british"  "chines"   "time"  
##  [3,] "world"  "budget"  "case"   "pay"     "london"   "offici"   "term"  
##  [4,] "peopl"  "billion" "legal"  "billion" "england"  "beij"     "like"  
##  [5,] "argu"   "year"    "right"  "fee"     "kingdom"  "hong"     "big"   
##  [6,] "think"  "deficit" "rule"   "year"    "unit"     "kong"     "run"   
##  [7,] "one"    "govern"  "judg"   "use"     "briton"   "shanghai" "also"  
##  [8,] "like"   "fiscal"  "lawyer" "paid"    "english"  "even"     "less"  
##  [9,] "good"   "will"    "suprem" "charg"   "scotland" "also"     "even"  
## [10,] "idea"   "tax"     "justic" "payment" "servic"   "yuan"     "well"  
##       Topic 65  Topic 98   Topic 34  Topic 56  Topic 1     Topic 35   
##  [1,] "measur"  "european" "law"     "histori" "space"     "road"     
##  [2,] "data"    "europ"    "rule"    "mani"    "one"       "line"     
##  [3,] "suggest" "union"    "regul"   "yet"     "mar"       "car"      
##  [4,] "also"    "countri"  "bill"    "war"     "scienc"    "train"    
##  [5,] "may"     "commiss"  "pass"    "day"     "technolog" "transport"
##  [6,] "can"     "nation"   "requir"  "never"   "field"     "rail"     
##  [7,] "chang"   "dutch"    "ban"     "father"  "light"     "railway"  
##  [8,] "base"    "brussel"  "new"     "centuri" "though"    "speed"    
##  [9,] "differ"  "germani"  "propos"  "old"     "earth"     "drive"    
## [10,] "point"   "franc"    "control" "much"    "orbit"     "say"      
##       Topic 73  Topic 12    Topic 28   Topic 2    Topic 82   Topic 58  
##  [1,] "will"    "project"   "protest"  "school"   "bond"     "women"   
##  [2,] "octob"   "build"     "street"   "educ"     "debt"     "children"
##  [3,] "month"   "plan"      "govern"   "student"  "rate"     "age"     
##  [4,] "next"    "will"      "call"     "univers"  "yield"    "famili"  
##  [5,] "novemb"  "water"     "polic"    "teacher"  "govern"   "young"   
##  [6,] "one"     "say"       "day"      "year"     "interest" "men"     
##  [7,] "year"    "river"     "demonstr" "colleg"   "market"   "old"     
##  [8,] "first"   "construct" "support"  "teach"    "investor" "parent"  
##  [9,] "said"    "built"     "peopl"    "children" "financ"   "sex"     
## [10,] "septemb" "billion"   "thousand" "pupil"    "borrow"   "child"   
##       Topic 86  Topic 70 Topic 97   Topic 19 Topic 95   Topic 4   
##  [1,] "say"     "reform" "polici"   "first"  "use"      "work"    
##  [2,] "mani"    "will"   "visit"    "second" "onlin"    "worker"  
##  [3,] "one"     "chang"  "also"     "back"   "data"     "job"     
##  [4,] "can"     "polici" "diplomat" "two"    "internet" "labour"  
##  [5,] "peopl"   "system" "leader"   "canada" "can"      "employ"  
##  [6,] "see"     "plan"   "presid"   "time"   "social"   "wage"    
##  [7,] "languag" "new"    "foreign"  "made"   "user"     "unemploy"
##  [8,] "still"   "need"   "two"      "place"  "applic"   "skill"   
##  [9,] "want"    "propos" "want"     "chang"  "search"   "young"   
## [10,] "main"    "polit"  "might"    "now"    "peopl"    "low"     
##       Topic 44  Topic 92  Topic 69     Topic 63 Topic 81   Topic 43
##  [1,] "countri" "time"    "state"      "chief"  "govern"   "peopl" 
##  [2,] "foreign" "test"    "unit"       "manag"  "state"    "kill"  
##  [3,] "world"   "ask"     "feder"      "boss"   "privat"   "fire"  
##  [4,] "global"  "experi"  "governor"   "execut" "public"   "mani"  
##  [5,] "intern"  "suggest" "california" "offic"  "sector"   "miss"  
##  [6,] "develop" "person"  "year"       "also"   "offici"   "now"   
##  [7,] "rich"    "control" "texa"       "job"    "own"      "bad"   
##  [8,] "emerg"   "word"    "san"        "board"  "say"      "least" 
##  [9,] "import"  "show"    "one"        "head"   "privatis" "disast"
## [10,] "abroad"  "think"   "counti"     "need"   "local"    "die"   
##       Topic 99 Topic 84       Topic 13   Topic 94     Topic 47   
##  [1,] "like"   "start"        "media"    "america"    "deal"     
##  [2,] "class"  "busi"         "televis"  "american"   "talk"     
##  [3,] "middl"  "firm"         "news"     "unit"       "negoti"   
##  [4,] "still"  "new"          "advertis" "state"      "agreement"
##  [5,] "well"   "entrepreneur" "newspap"  "washington" "two"      
##  [6,] "just"   "ventur"       "video"    "nation"     "agre"     
##  [7,] "much"   "small"        "show"     "long"       "sign"     
##  [8,] "better" "big"          "year"     "see"        "side"     
##  [9,] "half"   "founder"      "time"     "obama"      "want"     
## [10,] "part"   "capit"        "watch"    "action"     "will"     
##       Topic 88   Topic 89      Topic 75  Topic 18 Topic 59  Topic 45     
##  [1,] "secur"    "forc"        "africa"  "shop"   "poor"    "muslim"     
##  [2,] "agenc"    "militari"    "african" "retail" "peopl"   "egypt"      
##  [3,] "govern"   "defenc"      "south"   "store"  "help"    "islamist"   
##  [4,] "secret"   "armi"        "countri" "busi"   "poverti" "islam"      
##  [5,] "attack"   "arm"         "east"    "custom" "social"  "saudi"      
##  [6,] "say"      "war"         "kenya"   "sell"   "work"    "brother"    
##  [7,] "intellig" "afghanistan" "nigeria" "open"   "give"    "morsi"      
##  [8,] "offici"   "secur"       "middl"   "sale"   "incom"   "arab"       
##  [9,] "inform"   "soldier"     "region"  "onlin"  "benefit" "brotherhood"
## [10,] "spi"      "drone"       "say"     "say"    "money"   "arabia"     
##       Topic 32   Topic 10   Topic 36    Topic 9    Topic 71 Topic 7      
##  [1,] "hous"     "right"    "new"       "univers"  "music"  "climat"     
##  [2,] "price"    "group"    "york"      "paper"    "art"    "chang"      
##  [3,] "properti" "gay"      "one"       "research" "film"   "carbon"     
##  [4,] "home"     "campaign" "will"      "publish"  "one"    "warm"       
##  [5,] "land"     "marriag"  "year"      "public"   "show"   "environment"
##  [6,] "mortgag"  "support"  "old"       "book"     "man"    "tree"       
##  [7,] "rent"     "member"   "two"       "author"   "can"    "emiss"      
##  [8,] "new"      "among"    "bloomberg" "work"     "now"    "model"      
##  [9,] "valu"     "liber"    "boston"    "studi"    "cultur" "water"      
## [10,] "rise"     "debat"    "post"      "journal"  "first"  "global"     
##       Topic 8     Topic 14  Topic 79  Topic 87    Topic 51  Topic 42 
##  [1,] "asia"      "germani" "game"    "health"    "trade"   "immigr" 
##  [2,] "ship"      "german"  "sport"   "drug"      "market"  "mexico" 
##  [3,] "australia" "mrs"     "footbal" "hospit"    "exchang" "say"    
##  [4,] "island"    "merkel"  "club"    "care"      "import"  "migrant"
##  [5,] "south"     "green"   "play"    "treatment" "financi" "border" 
##  [6,] "sea"       "berlin"  "team"    "cancer"    "goldman" "year"   
##  [7,] "region"    "europ"   "world"   "patient"   "deriv"   "illeg"  
##  [8,] "port"      "social"  "player"  "doctor"    "econom"  "countri"
##  [9,] "countri"   "coalit"  "leagu"   "medic"     "world"   "home"   
## [10,] "indonesia" "left"    "can"     "also"      "financ"  "refuge" 
##       Topic 85 Topic 53   Topic 91  Topic 55 Topic 68     Topic 57
##  [1,] "food"   "power"    "oil"     "black"  "itali"      "tax"   
##  [2,] "say"    "energi"   "gas"     "mani"   "church"     "rais"  
##  [3,] "hotel"  "electr"   "billion" "white"  "italian"    "revenu"
##  [4,] "one"    "plant"    "energi"  "may"    "berlusconi" "pay"   
##  [5,] "meat"   "nuclear"  "price"   "like"   "left"       "incom" 
##  [6,] "drink"  "wind"     "reserv"  "race"   "christian"  "financ"
##  [7,] "eat"    "industri" "shale"   "make"   "one"        "rate"  
##  [8,] "good"   "generat"  "year"    "stop"   "cathol"     "govern"
##  [9,] "may"    "batteri"  "invest"  "less"   "movement"   "rich"  
## [10,] "consum" "renew"    "new"     "becom"  "hous"       "money" 
##       Topic 72  Topic 60   Topic 76     Topic 41  Topic 93    Topic 39 
##  [1,] "air"     "insur"    "brazil"     "pension" "french"    "russia" 
##  [2,] "new"     "health"   "year"       "pay"     "franc"     "russian"
##  [3,] "airport" "will"     "farmer"     "scheme"  "holland"   "putin"  
##  [4,] "fli"     "care"     "farm"       "public"  "pari"      "ukrain" 
##  [5,] "airlin"  "plan"     "latin"      "benefit" "fran"      "polit"  
##  [6,] "will"    "obamacar" "say"        "fund"    "mali"      "kremlin"
##  [7,] "flight"  "exchang"  "govern"     "will"    "left"      "moscow" 
##  [8,] "take"    "like"     "agricultur" "retir"   "socialist" "soviet" 
##  [9,] "dubai"   "cost"     "land"       "year"    "now"       "also"   
## [10,] "plane"   "mani"     "brazilian"  "detroit" "presid"    "now"    
##       Topic 100 Topic 37  Topic 38   Topic 3       Topic 6   Topic 17 
##  [1,] "mine"    "turkey"  "india"    "israel"      "japan"   "north"  
##  [2,] "make"    "erdogan" "indian"   "iran"        "abe"     "south"  
##  [3,] "gold"    "yet"     "say"      "isra"        "japanes" "korea"  
##  [4,] "world"   "say"     "state"    "palestinian" "new"     "park"   
##  [5,] "high"    "includ"  "year"     "arab"        "say"     "korean" 
##  [6,] "year"    "turkish" "delhi"    "west"        "tokyo"   "nuclear"
##  [7,] "wast"    "smoke"   "run"      "middl"       "minist"  "kim"    
##  [8,] "miner"   "may"     "congress" "east"        "countri" "regim"  
##  [9,] "steel"   "world"   "now"      "iranian"     "prime"   "state"  
## [10,] "littl"   "troubl"  "yet"      "state"       "now"     "test"

As one can see in the most common topics; economics, finance, business, politics top the list.

econ.3 = read.csv("/users/sweiss/google drive/economistdata.csv")
respons.vars = paste(1:100)
theta.econ = cbind(econ.3[, 1:3], theta)
date.theta.agg = aggregate(x = theta.econ[respons.vars], by = theta.econ["Date"], 
    FUN = mean)
date.theta.agg$Month <- factor(date.theta.agg$Date, levels = date.theta.agg$Date[!duplicated(date.theta.agg$Date)])
date.theta.agg[, 1] = as.Date(date.theta.agg[, 1])
library(scales)

library(ggplot2)
ggplot(data = date.theta.agg, aes(x = as.Date(Date), y = date.theta.agg[, 68])) + 
    geom_line() + geom_smooth() + scale_x_date(breaks = "months", labels = date_format("%b-%Y")) + 
    ggtitle("'Syrian Civil War' Topic include words: syria,rebel,war,assad,regim") + 
    xlab("Date (By Month)") + ylab("Average Percentage of Articles Attributed to Factor")

## geom_smooth: method="auto" and size of largest group is <1000, so using loess. Use 'method = x' to change the smoothing method.

Syrian Civil War 'peaked' in May/June. Since then the percentage of articles attributed to this topic has declined. After it became clear that USA, France, and UK were not going to intervene militarily, there were fewer articles written on the topic.

ggplot(data = date.theta.agg, aes(x = as.Date(Date), y = date.theta.agg[, 32])) + 
    geom_line() + geom_smooth() + scale_x_date(breaks = "months", labels = date_format("%b-%Y")) + 
    ggtitle("'Euro Crisis' Topic include words: euro,zone,spain,european,bail") + 
    xlab("Date (By Month)") + ylab("Average Percentage of Articles Attributed to Factor")

## geom_smooth: method="auto" and size of largest group is <1000, so using loess. Use 'method = x' to change the smoothing method.

Euro Crisis was with us the entire year…looks like this can continue on next year too!

ggplot(data = date.theta.agg, aes(x = as.Date(Date), y = date.theta.agg[, 29])) + 
    geom_line() + geom_smooth() + scale_x_date(breaks = "months", labels = date_format("%b-%Y")) + 
    ggtitle("'Protests' Topic include words: protest,street,govern,call,police") + 
    xlab("Date (By Month)") + ylab("Average Percentage of Articles Attributed to Factor")

## geom_smooth: method="auto" and size of largest group is <1000, so using loess. Use 'method = x' to change the smoothing method.

Protest topic was very high in June/July because of the Brazillian Protests and Turkish Protests. Sharp increase in final weeks of year of this topic from Thailand and even (gasp!) Singapore.

ggplot(data = date.theta.agg, aes(x = as.Date(Date), y = date.theta.agg[, 61])) + 
    geom_line() + geom_smooth() + scale_x_date(breaks = "months", labels = date_format("%b-%Y")) + 
    ggtitle("'Obama Care' Topic include words: insur,health,care,plan,obamaca") + 
    xlab("Date (By Month)") + ylab("Average Percentage of Articles Attributed to Factor")

## geom_smooth: method="auto" and size of largest group is <1000, so using loess. Use 'method = x' to change the smoothing method.

Obama Care topic increased in end of year due to botched introduction of healthcare website.

ggplot(data = date.theta.agg, aes(x = as.Date(Date), y = date.theta.agg[, 96])) + 
    geom_line() + geom_smooth() + scale_x_date(breaks = "months", labels = date_format("%b-%Y")) + 
    ggtitle("'Internet' Topic include words: use,online,data,internet,social") + 
    xlab("Date (By Month)") + ylab("Average Percentage of Articles Attributed to Factor")

## geom_smooth: method="auto" and size of largest group is <1000, so using loess. Use 'method = x' to change the smoothing method.

Finally, the Internet topic showed a strong decline ovoer the past year despite events like Twitter IPO. Perhaps this is an indication that social networks and internet are becoming so ubiquitious, its simply not news anymore.

Mideast Graph 2

I wasn't satisfied with my original plot  Previous Post of the Middle East situation a few months ago. After all, I just placed the Entities in an ad hoc way, and it showed no obvious structure despite my efforts...

Since then I've done some research and found out about a solution to find structure in networks with negative weights.(Check This Article...I followed the Example of Figure 6 ).

The structure becomes more apparent when we use this approach, which involves using two Eigen Vectors associated with the two smallest Eigen Values of the Laplacian Matrix.




























It's easier to see that Obama (US) and Turkey have similiar friends and enemeis. They do share a common enemy in General Sisi and common friend in Muslim Brotherhood. Iran by contrast supports Hamas and Assad who are against Obama and Turkey, so it makes sense that Iran is far away from Obama and turkey.

A hard look at the adjacency matrix could reach the same conclusions, but it's difficult for so many connections (this 8x8 matrix contains 28 different undirected relations). This method makes it easier to visualize the connections of friends and enemies alike.



Data
R Code

Why are Palestinians Killed by Israeli Defense Forces?

I watched 5 Broken Cameras and was surprised that Israeli Defense Forces (IDF) shot live ammunition at Palestinians protesting the security wall. One Palestinian Protester in the documentary was shot dead. I wanted to know how often protestors throwing stones were killed by the IDF. 

B'tselem has statistics for each fatality caused by IDF in both West Bank and Gaza Strip since 9/2000, with a short description of the event that took place. Some examples of these descriptions are show below:

2                          Killed while on his way to buy candy at a store next to her school.

3                                   Killed when Border Police came to his house to arrest him.

4                                                                         Killed in his house.

5                Killed during the arrest of his brother, who Wanted by Israel. Was not armed.

6  Wanted by Israel. Killed during an exchange of gunfire with soldiers who came to arrest him.

I'd like to classify these descriptions to see reasons people are killed by IDF. Doing so can give a better picture of the current conflict and problems to overcome.

First some descriptive Statistics:

6711 Total Observations - All considered Palestinian Citizens

517 Female and 6194 Males

Mean age is 25.38 years, with 19-29 1st-3rd quartile. 

Above is an Age Pyramid of Palestinians killed. Men are predominately the victims and have a slightly right skewed distribution with age. Women have a much more uniform distribution. 

Overall the demographics most likely to be killed by IDF forces are male and in their mid twenties.

Below is a time series plot of all Palestinians Killed monthly by Gaza or West Bank Location. 

Palestinians Killed in IDF confrontations Map

However, 539 deaths had no description.Below are those without time series:

Missing data appears to occur with high number of deaths. 

To classify the data I used Latent Dirichlet Allocation. This method assumes each text description or post is an amalgamation of independent topics. Each topic has a probability distribution over the set of words in all posts. For each description the algorithm takes the words as given and produces a probability distribution over the set of topics. In effect, each post will result in a combination of "Topics" and will be a sum of different these Topics. The benefit of LDA is that multiple topics can be applied to each description. 

Topics are displayed below in decreasing Importance:

Topic 7	Topic 9	Topic 4	Topic 14	Topic 5	Topic 12	Topic 19	Topic 1	Topic 15	Topic 18
polic	famili	gunfir	idf	fire	arm	want	gaza	oper	car
station	hous	exchang	incurs	rocket	men	israel	near	defens	ride
build	bomb	soldier	camp	street	area	arrest	fenc	day	anoth
near	member	kissufim	refuge	next	move	person	citi	shield	also
next	armi	attempt	rafah	school	walk	undercov	perimet	repent	drive
injur	tunnel	guard	jenin	mortar	present	unit	cross	rain	travel
council	warn	apart	albureij	qassam	asabra	came	strip	summer	road
legisl	use	farmland	provis	earlier	demonstr	action	northern	idf	man
abu	nine	nablus	yavn	hit	east	hide	east	katif	taxi
offic	refus	organ	build	jabalya	observ	attempt	approach	aim	two
Topic 8	Topic 16	Topic 10	Topic 11	Topic 20	Topic 6	Topic 2	Topic 17	Topic 13	Topic 3
way	idf	home	soldier	hous	tri	friend	person	palestinian	wound
attack	neighborhood	neighborhood	shot	stand	near	work	armi	forc	two
settlement	incurs	isra	stone	roof	border	home	bomb	secur	brother
tri	azaytun	beit	demonstr	entranc	explos	meter	hama	three	die
checkpoint	yuni	struck	throw	neighbor	assassin "area"	militari	along	later
carri	khan	shell	open	found	plant	sit	wing	incid	father
post	azeitun	missil	head	bodi	group	yard	mosqu	headquart	hospit
infiltr	locat	civilian	jeep	went	tank	land	store	four	peopl
soldier	pass	kill	part	mother	charg	two	islam	villag	injur
follow	check	hanun	petrol	son	place	agricultur	jihad	five	sister

Two Topics I thought were worth mentioning were topics 11 and 19.

Topic 11 includes terms "Wanted", "Arrest" "Israel", "Person", "Undercover". This topic discusses Palestinians killed by IDF soldiers who were wanted by Israel. 

Above is the average proportion of each post attributed to this topic. One can see that it increased in 2006 in West Bank, while remaining relatively constant in Gaza. Because Israel has much more control of WB than Gaza Strip, it makes sense that Israel tries and arrest more people in WB and results in more deaths as a result.

Topic 11 discusses those who died during demonstrations; the very topic that brought my interset tot he topic. Its interesting that there has been a steady increase in West Bank while Gaza has seen a slight decrease in these. The Separation Barrier could be the reason for this divergence in protesting.


R Code

Data