Those who have been keeping track of Bitnine’s work might be aware of the recently launched free cloud-based graph database service, AG Cloud Express. AG Cloud Express has AG Viewer included, which is a graph visualization tool that shows data in nodes and edges, enabling the user to draw visible insights much easier than staring long at a table.
At the beginning of the launch, four datasets were provided for the users to test AG Cloud Express. The main purpose of these datasets was to help new visitors understand how visualizing the graph database looks. However, ending it there would do no justice for cloud service with such great potential. After all, AG Cloud Express will keep adding new features until it grows up to be a full graph analytical enterprise service.
The first update of AG Cloud Express is a feature everyone has been waiting for.
You can now import your CSV files on AG Cloud Express.
If you tried out the new feature before reading the tutorial or this blog post, you might have encountered an error while importing your data. That is because there are certain requirements you must follow before importing your raw data into AG Cloud Express. Learn how to model your data and save your files in ways AG Cloud Express can understand.
If you have checked out the above tutorial, you will have noticed the three CSV files we prepared for example. The raw data of these files were acquired from Google Analytics. If you happen to manage your business’ data with GA, the tutorial will help convert your Google Analytics data into CSV files.
Analyzing Google Analytics data with AG Cloud Express
The three CSV files provided in the tutorial are as follows. We will import these files and see how it looks on AG Viewer, a graph visualization tool available within AG Cloud Express.
Region file shows data of regions that visited the online commerce website. This will be the start node.
Page file shows data of website pages. This will be the end node
This edge file shows the relationship between the two nodes. The edge includes properties that represent the number of visits made to the pages.
Open AG Viewer to see what these files look like in the graph visualization software.
Overall visits from many regions
As mentioned in a mock scenario below, we will take a look at what kind of insight we can get from this data.
A marketing specialist gathered a month’s worth of Google merchandise store data from Google Analytics. The specialist found out that California met the expected values of visits made to the websites, which would likely increase sales and achieve successful ROI. However, the specialist was not pleased to find out there was little success with the visitors from New York. The specialist brainstormed a way to increase traffic from New York, but to do so, a website that needed the most improvement had to be found.
By comparing New York to California, the specialist aimed to find out which pages both New York and California had in common and which pages did not.
After finding out such results, the specialist hoped to find which merchandise websites need improvement to target the visitors from New York. Since it will take some time to conclude the table data, the specialist would need to quickly visualize the data with AG Cloud Express.
To compare California and New York, the specialist has separately organized the table data of California and New York. The table data is preferred when seeing simple log data of one region.
Websites California visited
Websites New York visited
Next, the specialist imported the CSV files on AG Cloud Express to see which stores are in common with these two regions. In this case, the graph makes it easier to see the relationship between the regions and page sites. AG Viewer has a filter/search feature that allows you to highlight the two regions over the other regions. A bit of customization is needed to make graph data look clean as shown below.
= Filter / Search
At this point, it becomes clear which category of google merchandise store needs immediate attention from New York. All but the following stores had no connection with New York.
Shop by Brand
Google Cloud | Shop by Brand
Mens T-Shirts | Apparel
Android | Shop by Brand
YouTube | Shop by Brand
Hats | Apparel
Socks | Apparel
As shown in the table data, there are large gaps in the number of visits between the two states. With the help of our cloud service, the specialist was able to figure out which store needs immediate attention and start from there.
Another perk of AG Viewer is the feature to control the thickness of edges. Since the properties are included in edges, we can add weights to the edges, making the edges look thicker.
With it, the specialist can distinguish edges with the higher property. The specialist can distinguish which sites have more visits and determine why certain sites got more views than others through this feature. From this analysis, the specialist might be able to get a few resourceful insights that can greatly improve his/her digital marketing activity.
One of the perks of using a graph database is that you can view the relationships of multiple nodes. Comparing two states in a table will take longer to process, but it is still possible. But what if more data needs to be compared? If more states are added, then it will be difficult to rely on a relational database to make the final call for analysis.
With AgensGraph as a root, AG Cloud Express can help connect and see your data with a clearer vision. Do not hesitate to try our new graph-based cloud service now!
Connect your Data with AG Cloud Express
The purpose of the mock scenario was to show what AG Cloud Express is capable of. The newly added feature on AG Viewer can customize edge thickness, which can make one get a clearer insight from a simple layout. With AG Cloud Express, you can perform a significant level of analysis with simple data setting and query execution, and easily derive insights.
Thicker Edge shows higher visit
AG Cloud Express is still at an early stage of development, but thanks to AG Viewer, it can carry out basic functions such as finding relationships between multiple nodes and stacking properties from table data within nodes or edges (depending on the data modeling). For example, if you mouse over the edge between California and mens / unisex, you will find the event_total value is 14,232. The larger the value is, the bolder the edge will be displayed.
If you are a graph database expert with Cypher knowledge, you may be able to handle AG Cloud Express with ease since the Query Editor box of AG Viewer accepts only Cypher query language.
More advanced graph algorithms will be added, such as graph similarity, clustering, and shortest path. A deeper level of analysis will be possible when all these analytical algorithms are added to AG Cloud Express. Bitnine’s technology is proven to be known for improving countless business performances. The upcoming enterprise cloud-based graph visualization service, known as Cloud AG, will continue on Bitnine’s technology. Stay tuned until then and try out AG Cloud Express and get started with your data analysis now!
If you have any questions regarding the newly released AG Cloud Express, please feel free to reach us.
Thank you! 😀