| Column | DataType | SampleData |
|---|---|---|
| date_time | character | 10/10/1949 20:30 |
| city_area | character | san marcos |
| state | character | tx |
| country | character | us |
| ufo_shape | character | cylinder |
| encounter_length | numeric | 2700.00 |
| described_encounter_length | character | 45 minutes |
| description | character | This event took place in early fall around 1949-50. It occurred after a Boy Scout meeting in the Baptist Church. The Baptist Church sit |
| date_documented | character | 4/27/2004 |
| latitude | numeric | 29.88306 |
| longitude | numeric | -97.94111 |
Revealing Alien Patterns: A Data-Driven Investigation into Global UFO Sightings
Decoding the Mysteries: Analyzing Temporal, Spatial, and Descriptive Patterns in Over 80,000 UFO Sightings
Project Overview
The fascination with unidentified flying objects (UFOs) has captured the imagination of humanity for decades. This project aims to explore a comprehensive dataset comprising over 80,000 records of UFO sightings dating back to 1949. By leveraging the wealth of information provided, including latitude and longitude data, date and time stamps, and detailed descriptions of each sighting, we seek to unravel patterns, trends, and insights regarding UFO activities around the world.
Dataset Overview
Reasons for Choosing the Dataset:
The dataset under consideration comes originally from NUFORC contains over 80,000 records of UFO sightings dating back to 1949, making it a rich source for exploring the temporal and spatial patterns of extraterrestrial activities. The inclusion of latitude and longitude data, date and time stamps, and detailed descriptions of each sighting provides a comprehensive foundation for conducting a thorough analysis.
========== Structure of Dataset ==========
========== Overview of Dataset ==========
# A tibble: 11 × 6
variables types missing_count missing_percent unique_count unique_rate
<chr> <chr> <int> <dbl> <int> <dbl>
1 date_time char… 0 0 69586 0.866
2 city_area char… 0 0 19900 0.248
3 state char… 5797 7.22 68 0.000846
4 country char… 9670 12.0 6 0.0000747
5 ufo_shape char… 1932 2.41 30 0.000373
6 encounter_length nume… 3 0.00373 534 0.00665
7 described_encou… char… 0 0 8349 0.104
8 description char… 15 0.0187 79997 0.996
9 date_documented char… 0 0 317 0.00395
10 latitude nume… 1 0.00124 18421 0.229
11 longitude nume… 0 0 19455 0.242 ========== Dataset Dimensions ==========
Number of rows: 80332 Number of columns: 11 ======= First Few Rows of the Dataset ========
# A tibble: 6 × 11
date_time city_area state country ufo_shape encounter_length
<chr> <chr> <chr> <chr> <chr> <dbl>
1 10/10/1949 20:30 san marcos tx us cylinder 2700
2 10/10/1949 21:00 lackland afb tx <NA> light 7200
3 10/10/1955 17:00 chester (uk/england) <NA> gb circle 20
4 10/10/1956 21:00 edna tx us circle 20
5 10/10/1960 20:00 kaneohe hi us light 900
6 10/10/1961 19:00 bristol tn us sphere 300
# ℹ 5 more variables: described_encounter_length <chr>, description <chr>,
# date_documented <chr>, latitude <dbl>, longitude <dbl>Questions to Answer
Question 1:
How many UFO sightings have been reported worldwide, and are there particular regions where these sightings occur more frequently?
Question 2:
Is there any specific time of the day when UFO sighting are more prevalent?
Analysis plan
Data Wrangling : To prepare the dataset for analysis, several steps will be undertaken:
Data Cleaning:
Use functions like na.omit() to handle missing data.
Employ lubridate for standardizing date and time formats.
Feature Engineering:
- Utilize dplyr for creating a new variable representing the time of day by binning the existing date and time variable.
Filtering and Subsetting:
- Leverage dplyr to remove irrelevant columns and subset the data based on the desired time range.
Geocoding:
- If necessary, use packages like ggmap for geocoding to convert location names into latitude and longitude coordinates.
Exploratory Data Analysis (EDA):
- Leverage functions like summary() and plot() to perform initial exploratory data analysis, identifying outliers and gaining insights.
Normalization:
- Normalize data if required, especially if there are variations in the reporting of sightings over the years.
Analysis Plan for Q1 :
Plan: We plan to utilize latitude, longitude & State data to visualize the global distribution of UFO sightings, and identify geographical hot-spots and trends to understand if there are regions more prone to UFO activities.
We will emphasize this correlation in our presentation, dedicating a section to exploring and highlighting any relationships between population density and the frequency of UFO sightings. The specifics of this correlation will become clearer as we progress with the analysis.
Type of Plot: Heatmap or Choropleth Map
Rationale: Using packages like ggplot2 and leaflet, a geographical heatmap or spatial point density plot can effectively showcase the concentration of UFO sightings across different regions, providing a clear visual representation of hotspots and trends.
Variables Involved: Latitude, Longitude & State
Additional Statistical Approaches for Q1 :
Descriptive Statistics: We will provide summary statistics such as the mean, median, and standard deviation of the number of sightings globally. This will give us an overall understanding of the distribution of sightings.
Geospatial Clustering: We will use clustering algorithms, such as K-means, to identify regions with a high concentration of sightings. This can help identify specific geographic areas where UFO sightings are more prevalent.
Spatial Autocorrelation: We will assess spatial autocorrelation to determine if there are spatial patterns in the distribution of sightings. This can be done using Moran’s I index.
Analysis Plan for Q2 :
Plan: R offers packages such as ggplot2 for time series plotting. A time series plot or bar chart can visually represent temporal trends in UFO sightings throughout the day, helping identify patterns associated with different times of day.
Type of Plot: Line Plot or Bar Chart.
Rationale: We can use a line plot or bar chart to illustrate the temporal trends in UFO sightings throughout the day. This visual representation will help identify peak hours and potential patterns associated with different times of the day.
Variables Involved: Time of Day (New variable for time of day binning to be build using date_time variable).Will extract time from ‘date_time’ variable and create time bins (e.g., morning, afternoon, evening, night) and analyze sighting patterns within each bin.
Morning:
Start time: 6:00 AM End time: 12:00 PM (noon)
Afternoon:
Start time: 12:00 PM (noon) End time: 6:00 PM
Evening:
Start time: 6:00 PM End time: 10:00 PM
Night:
Start time: 10:00 PM End time: 5:59 AM (next day)
Additional Statistical Approaches for Q2 :
Density plot to visualize the hourly distribution of UFO sightings. This will provide a more granular view of when sightings occur within each hour.
Conduct statistical tests (e.g., chi-square test) to determine if there are significant differences in sighting frequencies between different time periods (morning, afternoon, evening, night).
Conclusion
This project seeks to bring a data-driven approach to the mysterious realm of UFO sightings. By analyzing historical data, we aim to uncover patterns, answer questions about temporal and spatial characteristics, and contribute to our understanding of these extraterrestrial phenomena. The findings may have public awareness, providing valuable insights into the unknown.
Additional Points :
Impact on Public Perception:
This highlights how the findings of this project can impact public perception and awareness of UFO sightings. The analysis can shed light on the frequency and distribution of sightings, potentially influencing how people perceive these phenomena.
Scientific Inquiry and Exploration:
This emphasizes the importance of using a data-driven approach to explore UFO sightings. This project contributes to scientific inquiry by applying rigorous analysis techniques to historical data, fostering a better understanding of these mysterious occurrences.
Potential for Further Research:
This discusses how the insights gained from this project can serve as a foundation for future research in the field of ufology and related disciplines. The patterns and trends uncovered may spark new questions and avenues for exploration within the scientific community.
Public Engagement and Education:
It highlights the potential for public engagement and education regarding UFO sightings. The project’s findings can be disseminated through various channels, including educational programs, documentaries, and public seminars, to increase awareness and understanding of these phenomena.
Timeline
| Serial No. | Tasks | Responsibility | Expected Time of Completion |
|---|---|---|---|
| 1 | Data Wrangling | Tushar, Ajay & Eeshaan | 26th Feb’24 |
| 2 | Analysis of Question 1 | Ajay & Maheedhar | 4th March’24 |
| 3 | Analysis of Question 2 | Ajay & Nandini | 8th March’24 |
| 4 | Finalizing Write Up, Presentation Slides & Project Website | Alexandar, Tushar, Ajay, Eeshaan, Nandini & Maheedhar | 11th March’24 |