SUMMARY

The quantified self dashboard is an effort to learn and analyze a few, if not all, aspects of my life. In this dashboard I have tried to highlight and answer five questions. In the process of answering these questions, I learned a great deal about myself. Since coming to the United States of America, I have tried making the most of my free time when I am not working or studying. My focus on traveling, exploring TV shows, indulging in leisurely/entertainment activities, and taking care of my health were all an effort to grow as a whole and be better than I was yesterday.

In the visualizations presented, I focused on, first, the cities I travelled and lived in US since 2014; second, days I watched Netflix the most since 2016; third, extending the Netflix analysis, highlighting the shows I watched the most; fourth, a visualization showing my spending habits; lastly, an interactive chart of the daily steps I took since January 2018, an insight into my efforts to stay active.

DASHBOARD

Column1

CITIES VISITED IN USA

NETFLIX WATCH TIME

NETFLIX WORDCLOUD

CREDIT CARD ACTIVITY

DAILY ACTIVITY SINCE JAN’18

REPORT

DATA COLLECTION PROCESS

For this dashboard, I had to look at a few places to get the data. For the first graphic I recalled and listed the cities I had lived and visited in the past, and then using the function “geocode” in R recorded the longitudes and latitudes. Next, I scrapped my Netflix watching history from my Netflix account, later I used this data to, one, analyze the days I used my account, and second, which shows I had watched the most. After this, I collected data from my credit card statements that categorized each activity. Lastly, I gathered data from a mobile app I use that tracks the steps taken every day. All of the data was stored and managed in a .csv file so I can create the visualizations.

ANALYSIS

Figure 1: Using “ggmap” and “geocode”, I created the cities I have visited and traveled. Using the density points, I was able to highlight that Syracuse, Jacksonville, and New York City are the cities I lived in. It was also interesting for me to see that I have not explored the West Coast at all, and need to plan a few trips to that part of the world soon.
Figure 2: Since the watching history on Netflix was time stamped, I thought of digging deep to see which days I watched the most TV. It was quite Intuitive that on Sundays I saw the most shows since I was mostly at home. It was interesting to see that Tuesdays and Wednesdays had the least number of titles watched since on those days I have online classes at Harrisburg after work.
Figure 3: The wordcloud showed the TV shows I can watch again and again. While “Friends”, “The Office”, “Forensic Files”, “Malcom in the Middle”, “Archer”, “Supernatural” are all time favorites, I also explored other shows such as “House of Cards”. The graphic also brought out the genres I like in general.
Figure 4: I created a bar chart by categories for my credit card expenses. Since my major expenses are on flights and hotels because I like to travel, this emerged as the biggest category. On the contrary, Supermarkets (I seldom cook at home) and Gasoline (because I don’t have a car yet, and don’t rent that often) were among the smallest categories.
Figure 5: I used “dygraphs” to generate a time-series chart to summarize the steps I have taken since January 2018. It was quite fascinating to see that I averaged about 8,000 steps a day - an impressive feat for a couch potato. The interactive feature of Range-Selector let me focus on specific periods during this time and analyze if I burned enough calories.

CONCLUSION

In conclusion, this dashboard has really opened my eyes as to where I spend most of my energy, money, and time. Going forward, this will definitely help me track and utilize the resources I have in a better way. Thank you for taking some time to take a peak into my life.

---
title: "THE QUANTIFIED SELF"
author: "Himanshu Soni"
output: 
  flexdashboard::flex_dashboard:
    orientation: columns
    vertical_layout: fill
    social: menu
    source: embed
---

```{r setup, include=FALSE}
library(readxl)
library(ggplot2)
library(ggmap)
library(NLP)
library(tm)
library(SnowballC)
library(RColorBrewer)
library(wordcloud)
library(dygraphs)
library(xts)
```

#SUMMARY {.sidebar}
The quantified self dashboard is an effort to learn and analyze a few, if not all, aspects of my life. In this dashboard I have tried to highlight and answer five questions. In the process of answering these questions, I learned a great deal about myself. Since coming to the United States of America, I have tried making the most of my free time when I am not working or studying. My focus on traveling, exploring TV shows, indulging in leisurely/entertainment activities, and taking care of my health were all an effort to grow as a whole and be better than I was yesterday.

In the visualizations presented, I focused on, first, the cities I travelled and lived in US since 2014; second, days I watched Netflix the most since 2016; third, extending the Netflix analysis, highlighting the shows I watched the most; fourth, a visualization showing my spending habits; lastly, an interactive chart of the daily steps I took since January 2018, an insight into my efforts to stay active.

#DASHBOARD {data-icon="fa-map"}

##Column1 {.tabset .tabset-fade}
### CITIES VISITED IN USA

```{r CitiesVisited, echo=FALSE}
US_map <- qmap('Dallas', zoom = 4, source = "google", maptype = "roadmap", scale = 2, extent = "panel")
CitiesVisited <- read_xlsx("/Users/Himanshu/Desktop/Cities.xlsx", sheet=1)
US_map + geom_point(aes(x=lon, y=lat), data = CitiesVisited, col = "dark blue", alpha = 0.5, size = CitiesVisited$Stay) + labs(title="Number Of Cities Visited In USA", x="Longitude", y= "Latitude")
```

### NETFLIX WATCH TIME

```{r Netflix, echo=FALSE}
Netflix <- read_xlsx("/Users/Himanshu/Desktop/Cities.xlsx", sheet=2)
Netflix$Day <- factor(Netflix$Day, levels = c("Monday","Tuesday","Wednesday","Thursday","Friday","Saturday","Sunday"))
ggplot(data = Netflix, aes(x=Day, fill = Day)) + geom_bar(stat = 'count') + scale_fill_brewer() + labs(title="Number Of Titles Watched On A Given Day", x="Day Of The Week", y= "# Of Titles") + theme(legend.position = "right")

```

### NETFLIX WORDCLOUD

```{r}
Docs<- Corpus(DirSource("/Users/Himanshu/Desktop/Netflixfiles"))
tdm <- TermDocumentMatrix(Docs,control = list(removePunctuation = TRUE,stopwords = c("the", "season","one", stopwords("english")), removeNumbers = TRUE, tolower = TRUE, stripWhitespace = TRUE))
m = as.matrix(tdm)
word_freqs = sort(rowSums(m), decreasing=TRUE)
dm = data.frame(word=names(word_freqs), freq=word_freqs)
wordcloud(dm$word, dm$freq, random.order=FALSE, colors=brewer.pal(8, "BrBG"))
```

### CREDIT CARD ACTIVITY

```{r Credit, echo=FALSE}
Categories <- read_xlsx("/Users/Himanshu/Desktop/Cities.xlsx", sheet=3)
Categories$Category <- factor(Categories$Category, levels = c("Travel/Leisure", "Restaurants", "Merchandise", "Supermarkets", "Services", "Gasoline"))
ggplot(data = Categories, aes(x=Category, fill = Category)) + geom_bar(stat = 'count') + scale_fill_brewer() + labs(title="Top Categories Of Expense By Number Of Transactions", x="Expense Categories", y= "# Of Transactions")  + theme(legend.position = "bottom")
```

### DAILY ACTIVITY SINCE JAN'18 {data-padding=50}
```{r Activity, echo=FALSE}
Activity <- read_xlsx("/Users/Himanshu/Desktop/Cities.xlsx", sheet=4)
Activity$ActivityDate <- as.Date(Activity$ActivityDate, format = "%m/%d/%Y")
time_series <- xts(Activity, order.by = Activity$ActivityDate)
dygraph(time_series) %>% dyRangeSelector()
```

#REPORT {data-icon="fa-list"}

**DATA COLLECTION PROCESS**

For this dashboard, I had to look at a few places to get the data. For the first graphic I recalled and listed the cities I had lived and visited in the past, and then using the function "geocode" in R recorded the longitudes and latitudes. Next, I scrapped my Netflix watching history from my Netflix account, later I used this data to, one, analyze the days I used my account, and second, which shows I had watched the most. After this, I collected data from my credit card statements that categorized  each activity. Lastly, I gathered data from a mobile app I use that tracks the steps taken every day. All of the data was stored and managed in a .csv file so I can create the visualizations. 

**ANALYSIS**

- *Figure 1:* Using "ggmap" and "geocode", I created the cities I have visited and traveled. Using the density points, I was able to highlight that Syracuse, Jacksonville, and New York City are the cities I lived in. It was also interesting for me to see that I have not explored the West Coast at all, and need to plan a few trips to that part of the world soon. 

- *Figure 2:* Since the watching history on Netflix was time stamped, I thought of digging deep to see which days I watched the most TV. It was quite Intuitive that on Sundays I saw the most shows since I was mostly at home. It was interesting to see that Tuesdays and Wednesdays had the least number of titles watched since on those days I have online classes at Harrisburg after work. 

- *Figure 3:* The wordcloud showed the TV shows I can watch again and again. While "Friends", "The Office", "Forensic Files", "Malcom in the Middle", "Archer", "Supernatural" are all time favorites, I also explored other shows such as "House of Cards". The graphic also brought out the genres I like in general.

- *Figure 4:* I created a bar chart by categories for my credit card expenses. Since my major expenses are on flights and hotels because I like to travel, this emerged as the biggest category. On the contrary, Supermarkets (I seldom cook at home) and Gasoline (because I don't have a car yet, and don't rent that often) were among the smallest categories.

- *Figure 5:* I used "dygraphs" to generate a time-series chart to summarize the steps I have taken since January 2018. It was quite fascinating to see that I averaged about 8,000 steps a day - an impressive feat for a couch potato. The interactive feature of Range-Selector let me focus on specific periods during this time and analyze if I burned enough calories. 

**CONCLUSION**

In conclusion, this dashboard has really opened my eyes as to where I spend most of my energy, money, and time. Going forward, this will definitely help me track and utilize the resources I have in a better way. Thank you for taking some time to take a peak into my life.