Skip to main content

Contemporary Applications of Big Data in Science

 Healthcare - Data science used the data to identify and predict disease to personalise any healthcare recommendations. In 2008 Google created ‘Google Flu Trends’ which helped to map flu outbreaks by tracking location data on confirmed cases of the flu. But unfortunately it wasn’t successful. But it still showed the importance of data science in healthcare. Other examples are that Google developed a tool called LYNA that was used to identify breast cancer tumours because they can be difficult for people to see with just their eyes, this successfully identified nodes 99% of the time however it needs further testing before being able to be used by doctors in hospitals. Tracking periods by using apps such as ‘Clue’ to predict users menstrual cycles and also reproductive health by using things such as last start date, moods, stool type, hair condition and other different metrics. 

Sports - ‘Strong Analytics’ uses pre recorded footage to create detailed plan to help coaches to improve athlete performance. ‘WHOOP’ makes devices that athletes can wear to track their resting heart beat, sleep cycle, respiratory rate etc. because if this athletes can see when they should be pushing themselves at training and when they should be resting because of this they can make the most efficient steps to get the most out of their body and their fitness. 

https://builtin.com/data-science/data-science-applications-examples


Comments

Post a Comment

Popular posts from this blog

Limitations of Predictive Analysis

One of the limitations is data quality, the predictive models rely entirely on large and accurate datasets to make realistic predictions. So if the data is inaccurate, biased or incomplete then the model will be flawed.  The next is overfitting which happens when a model is trained on a particular dataset and becomes too complicated makes it difficult to generalise new data. Because of this it can make inaccurate predictions and poor performance.  Next is changing conditions, predictive analytics models are made to predict future outcomes based on previous data. However, the future is very uncertain and because of this the conditions can change quickly so it can be very difficult to predict accurately.         And lastly ethical concerns can make it difficult to predict analysis because of ethical concerns such as bias or privacy. Predictive models can have an already existing bias and discrimination. Also predictive analytics can have privacy concerns...
Post a Comment
Read more

Data Mining Methods

Data mining is the process of using statistical analysis and machine learning to reveal hidden patterns or odd things in large datasets because of this you can help make important decisions and predict what’s going to happen. it’s where you take data such as structured data, an image, video, text etc and train it, deploy and serve it then you can get actionable insights and application events.  The techniques are classification which is used to organise data into different classes or categories so it trains a model on labelled data and uses it to predict the class. Regression is used to predict numbers or continuous values based on relationships so it finds the function or model that best fits the data to make accurate predictions.  Clustering is used to group similar data uncover patterns or structures in the data without any classes or labels.  There is also association rule, anomaly detention, time series analysis, neural networks, decision trees, ensemble methods and ...
Post a Comment
Read more

Future Applications of Big Data

 There’s several ways in which the Big data in future applications some of which are: The increasing volume of big data analytics which in the future will focus on the freshness of the data with real time analysis which helps to make better decisions which have more information and additionally increase competition. It also is essential for gaining real life vision however can have some implications because of higher risk of having inaccurate or incomplete data. E.g ‘Snowflake’ announced ‘Snowpipe’ they republished their Kafka connector which made it so that when data lands in Snowflake it is immediately questioned which results in a 10x lower latency.  Being able to see the data that’s happening right this second so things like trading bitcoin etc. Because of this is that it has shaken up industries such as finance and social media. For example Walmart has created one of the largest hybrid clouds to help manage their supply chains and analyse sales n that moment and not a eee...
Post a Comment
Read more