Big Data
- Created by: miacampbell31
- Created on: 29-04-22 11:39
View mindmap
- Big Data
- Big data is a term associated with data sets thatare so complex that traditional database and other processing applications areunable to capture, curate, manage andprocess them within an acceptable time frame.
- Big data challenges can be defined as the 3V's:
- Volume - the amount of data to be processed.
- Variety - the number of types of data to be analysed.
- Velocity - the speed of data processing.
- Big data can be defined as, "high volume, high velocity, and/or high variety information assets that require new forms of processing to enable enhanced decision making".
- Social media is one of the biggest sources of big data. Consumer good companies actively scan social media websites to decipher user preferences, choices and perceptions towards their brands.
- Data sources can be categorised into internal and external. The internal data includes sources such as customer details, product details, sales data etc.
- External sources include data collected from business partners, data suppliers, internet, government and market research companies.
- commonly used data sources are: Social Media Machine data Transactional data
- Machine data is data generated from devices such as RFID chip readers, GPS results.
- Transactional data is data generated from companies such as eBay, Amazon and large stores such as Tesco's.
- Big data processing techniques analyse data setsat terabyte or even petabyte scale. Some methods of processing applied to big data include:
- Cluster analysis. Where groups of data records are identified.
- Classification. Where the data mining process is used to determine an appropriate structure to new data, in the way for example an email application may classify some emails as spam.
- Anomaly detection. Where unusual records are identified. Such anomalies may merit further investigation as ap oint of interest to the organisation or they may be representative of data errors.
- Regression. Where relationships between data variables are investigated to help how a change in an independent variable can impact upon a dependant data variable.
- Summerisation. Where data is summarised in a visual format.
- Regression. Where relationships between data variables are investigated to help how a change in an independent variable can impact upon a dependant data variable.
- Association rule mining and sequential pattern mining. Where dependencies between data itmes can be identified, for example the use of data sets by a supermarket to determine which patterns of products are pruchased together.
- Anomaly detection. Where unusual records are identified. Such anomalies may merit further investigation as ap oint of interest to the organisation or they may be representative of data errors.
- Classification. Where the data mining process is used to determine an appropriate structure to new data, in the way for example an email application may classify some emails as spam.
- Cluster analysis. Where groups of data records are identified.
- How the financial services sector uses big data.
- Ensuring they are complying with regulations. Using tradional data processing platforms to support this objective has become increasingly expensive and unsustainable Complex fuzzy matching can be applied on a name matching and contact information across much larger data sets at a lower cost.
- Improving risk analysis. Complex algorithms can be run on large volumes of transaction data to help identify fradulent activity or to perform risk analysis.
- Understanding customer behaviour and transaction patterns. Customer data can be consolidated from a variety of sources and analysed to predict customer spending, mortgage defaults.
- Imrpoving services. Customer data can be analysed to help identify patterns which can lead to customer dissatisfactionEmail content, telephone call recordings and social media comments can be analysed to determine postive/negative feelings of customers towards products on offer by the organisation.
- How the health sector uses big data.
- Used to predict epidemics, cure disease, improve quality of life and avoid preventable deaths.
- Some health organisations take data from various sources and use it to tailor health care programs for their clients.
- Such a high volume of data can also beanalysed alongside other clients to help identifyhealth patterns and threats using sophisticatedmodelling processes while on an individual level thetreatment programmes can be developed based onreliable and real-time data collected with regardsthe patients genetic makeup and lifestyle.
- How the retail sector uses big data.
- Used to predict trends and forcasting demand. Retails today make use of a wide range of data to help identify trends in products.
- Price optimisation. Once there is an understanding of the products people are interested in buying, retailers are able to use big data to determine where the demand will be.
- Through the analysis of demographic dataand economic indicators, spending habits ofcustomers can be identified while algorithmswhich track millions of transactions every daycan be used to track demand against inventoryand competitor activity to ensure a retailer canrespond quickly to real time changes in marketactivity.
- Identifying potential customers. Data collected through transactional records and loyalty programs allows demand to be forecast on the basis of geographic areas.
Comments
No comments have yet been made