How to choose the Right data analysis tool

Choose a good data analysis tool, you must understand the analysis of what data, big data to analyze the data types are mainly four categories:

1 , transaction data (TRANSACTION)

The Big data platform is able to capture a larger and larger amount of structured transaction data, so that more extensive transaction data types can be analyzed, not just pos or ecommerce shopping data, but also behavioral transactional data, such as Internet clickstream data logs from Web server records.

2 , human data (human-generated)

Unstructured data is ubiquitous in e-mail, documents, pictures, audio, video, and streams of data generated through blogs, wikis, and especially social media. This data provides a rich source of data for analysis using the Text Analytics feature.

3 , mobile data

Smartphones and tablets that are able to surf the web are becoming more common. Apps on these mobile devices are able to track and communicate countless events, from trading data within the app (such as recording events for search products) to personal information or status reporting events (such as location changes that report a new geocoding).

4 , machine and sensor data

This includes data created or generated by functional devices, such as smart meters, smart temperature controllers, factory machines, and household appliances connected to the Internet. These devices can be configured to communicate with other nodes in the internetwork and can automatically transfer data to a central server so that data can be analyzed. Machine and sensor data are the main examples from the emerging internet of Things (IoT). Data from the Internet of things can be used to build analytic models that continuously monitor predictive behavior, such as when a sensor value indicates a problem, and provide prescribed instructions (such as alerting technicians to check the device before a real issue occurs).

However, what are the requirements and objectives of the data analysis tools? Can apply advanced analysis algorithms and models to provide analysis, big data platform for the engine, such as Hadoop or other high-performance analysis system, can be applied to a variety of data sources of structured and unstructured data, as the data for the analysis model increases, can be extended, Analytical models can, or have been integrated into, data visualization tools and can be integrated with other technologies, and tools must include some of the necessary features, including integrated algorithms and support for data mining techniques, including (but not limited to):

(1) Clusters and subdivisions: divide a large entity into small groups with common characteristics. For example, analysis of the collected customers, to identify a more segmented target market.

Classification: organize the data into predetermined categories. For example, according to the subdivision model to determine how the customer classification.

(2) Recovery: used to restore dependent variables and relationships between one and more independent variables to help determine how dependent variables change depending on the independent variable. For example, use geographic data, net income, summer average temperature and floor space to predict the future trend of property.

(3) Joint and project collection mining: finding correlations between variables in a large data set. For example, it can help call center representatives to provide more accurate information based on the caller's customer segments, relationships, and complaint types.

(4) Similarity and linkage: used for the non-direct clustering algorithm. The similarity integration algorithm can be used to determine the similarity of entities in an alternate cluster.

(5) Neural Networks: non-direct analysis for machine learning.

How to choose the Right data analysis tool