Please enable JavaScript.

Coggle requires JavaScript to display documents.

F201 TA2 - Coggle Diagram

- - - - Is when you have a licences that allows the user to download and edit a source code .
      - Can be used when you need to tailor the software to your needs . You can change what the software does , add to it or remove elements etc making it efficient for your purpose
      - Is downloaded from the internet so there could have been edit by anyone meaning you need to be careful in case it contains malware .
    - - Is created by a company to sell it
      - This is a good choice when you want software that has been thoroughly tested before it is released .
      - Without the source code what the software can do is fixed how may be able to tailor some elements but if the software does not do everything you need you cannot change it .
    - - Converts data into visual styles e.g. graphs and charts This software takes data provided and converts it into a human friendly form .
      - Instead of reading pages of numeric data , the software can create a graph to enable people to easily see the changes or patterns in data
      - Not all data can be represented or a graph . Some data that can be represented as a graph must be cleaned in order to be shown in this form .
    - - Can be used to perform the processes to convert the data into meaningful data .
      - Can be used when you want perform calculation and processes on sets of data to produce results .
      - Some data analytics software will require the data to be in a specific format for analysis .
    - - Stores data in a non-tabular formats such as docs or graphs , offering a flexible structure does not rely on predefined tables or fields like SQL databases do .
      - Are highly scalable and can handle a wide variety of data types making them suitable for diverse and evolving data needs .
      - Performing structured queries is more complex in no sql databases because the data is not uniformly defined unlike sql databases where fields and relationships are clearly structured
    - - Is a software that allows data from many different places to be combined into one system . The characteristics are security , measures for maintaining data integrity , audit loss and scalability .
      - A data integration platform can be used to collect the data from the different sources and organise it into one set of data e.g removing duplicates and marking sure it is all in the same format .
      - They need to erasure the security of the data when it is stored and during transfer .
  - - - Is owned and run by one person or organisation
      - Allows the organisation to fully use all the hardware ; it is not sharing access , so it has full bandwidth access . The server can be setup for the organisation so that it fully meets its requirements
      - Can be expensive to set up and maintain . If it is run an organisation then it needs to employ people to manage the server and ensure it is always functions
    - - Are when multiple servers are used for one organisation over multiple locations .
      - Allow for one server to fail and other servers to reman accessible
      - Include high cost of purchasing , maintaining and running the servers
  - - - Is provided an organisation that offers cloud storage to anyone who wants to use it .
    - - Is a cloud storage that is only accessible to an organisation mainly used for sensitive data .
    - - Is a central system where data is stored in a structured way .
    - - Is a central systems where data is stored and can be analysed data is stored on the data lake in its unprocessed form from a range of sources differences between data lake and data warehouse .
    - - Is a type of secondary storage device is a hardware that stores data long term and it external therefore keeping data even after it has been turned off.
      - Large storage capacity , small , robust and fast data access
      - High cost per unit of storage capacity and limited life span.
    - - Is a decentralised (data that is not stored in one place ) data storage technique .
      - Data cannot be changed without authorisation as the data is stored in multiple locations , the data is decentralised reducing the risk of data loss and theft and transactional are secure due to encryption .
      - Slow access to data , not easily stable because of the storage space needed to store the same data multiple times and data takes longer to store and retrieve because of the processes to create a block chain .
    - - DNA refers to the genetic instructions that create organisms (humans)
      - Large amount of data stored in a small space and the storage is long lasting
      - Current lost are extremely high making it unrealistic for common use ( specialist equipment etc) and currently very slow storage and access .
- - - - Identify potential fraud asks for verification and identifies whether people should be given credit if at risk .
    - - Predict ways to increase customers and suggest which items be most successful .
    - - Predict which students will need more support and predict which students are at risk
    - - Predict ways to increase customer range and suggest which items will be most successful e.g. what types of game people are more likely to play.
    - - Predict where energy use my increase and predict customer who may not pay for their utilities
    - - Predict where additional resources will be needed and predict population changes .
    - - Predict which people will be at risk to an illness and predict responses to treatments.
    - - Predict the risk of a customer if they can be insured and predict customers who are likely to raise claim .
    - - Predict which materials are going to increase or decrease in cost and predict demand for products
  - - - Provide info about a customers past spending and provides info about previous transactions
    - - Identify which products have sold the most and identify the characteristics if customers who have purchased items .
    - - Identify which programs were watched and identify user preferences for each demographic .
    - - Identify students who are under achieve or overachieve and identify students attendance .
    - - Identify energy used in a given time by a customer in an area and identify which customers do not pay a time .
    - - Identify which areas more the lowest quality of living and identify were more money needs to be invested .
    - - Identify the number of patients with an illness over a year .
    - - Identify number of client claims over the year and identify the target of claims made by customers
    - - Identify the requirements of each product created and sold over the past year and identify of changes in requirements form customers .
  - - - Is used to idenifty if there is a relationship between two or more different variables . The variables can be visualised or by a graph with the results marked where they intersect
      - Can identify the type of relationship between variables , identify strengths and can be simpler to create algorithms
      - An outlier can significantly impact the results and can only work with a small amount of variables
    - - Works out the probability of something happening and it also follows for a level of randomness . The randomness is something that could happened and affect the outcome
      - Allows multiple possible outcomes to be considered ; most possibilities considered the more likely prediction to be accurate and to can produce probabilities of events occurring so the chances can be compared.
      - The prediction is not necessarily going to be accurate ; not all possibilities can be taken into account and if one extreme occurs it will be outside the predication. The analysis can be difficult to create and can be computationally intensive when the models are complicated
    - - It creates a smaller data set that is easier to analyse
      - Takes variables and reduces them , it identifies similarities and patterns between variables to create a smaller data model .
      - There needs to be a strong relationship between the variables to produce accurate groupings and the groupings may not be fully representations of each variable .
    - - Is to spilt the data gathered into a different data set for each cohort . Works well with data that can be grouped easily
      - Useful when results are needed for different groups e.g e-commerce
      - Long term cohort analysis can be difficult
    - - Is used to identify groups of data . Data can be represented visually on a graph similar to regression algorithms the positions of the mapped points can then be grouped according to which are closets to each other
      - It can group data together and produce results that generate behaviour .
      - There are always outliers that may not fit into a group or data may not lend itself to groupings
    - - Considers data over time . A time series is a representation of time with the data mapped at each point.
      - Changes can be seen over time , can provide an understanding of when changes have occurred and can be used to predict future changes .
      - Only shows the changes not the events that have happened to make these changes take place and can be affected by outliers without explanation of where or why theses occurred