As reported in Forbes, businesses are really starting to wake up to the promise of what we call Enterprise AI. But what does this mean for a non-coding analyst?

‍

‍

Exponentially increasing insights from data is critical to the transition to Enterprise AI, and it's a matter of scale — using more available data faster for more data projects. This cannot happen without expanding the scope of people who have access to and work with data on a daily basis.

Still, incorporating the work of non-data scientists into data projects in meaningful ways requires a fundamental change of mindset about data tools, namely the addition of powerful features that allow data analysts to do effective work without coding.

Dataiku offers powerful features in this regard; through the visual analysis layer, data analysts can easily discover, prepare, enrich and visualize various types of structured and unstructured data.

Here are just seven of the most important no-code features Dataiku offers that will change the way you work with data;

1. Intelligent Data Acquisition (Data Ingestions)

Contact IT, request access to data, wait for days (maybe weeks), gain access, clean up data, repeat. Does it sound familiar? There is a better way for Dataiku to be the solution to this long process; install datasets to work with XLSX or CSV files or connect to other resources (databases, files hosted on the server, links to business applications, etc.).

And Dataiku offers smart solutions. For example, when you upload a CSV or Excel file, it automatically recognizes separators and character encoding. It also displays a preview of the dataset for you to check if everything is in order, including the number of rows, and allows you to modify some parameters (row skipping, handling column headers, etc.) to make sure the data is correct.

Other useful no-code smart data retrieval features include:

• Merge all size, shape, and location data without entering million-row constraint
• Automatically collect the same files into a single dataset by dragging and dropping them to the interface
• Combine spreadsheet tabs into a single dataset
• Rename columns and set data types directly in the schema panel so that specific columns are always stored in certain formats

‍

2. Working with Dates and Times

Time-based features are commonplace in data-driven use cases and can be really challenging to work with. Depending on the original format, you may need to heavily recode to parse dates into a recognized date format.

In Dataiku, the smart date processor will recognize possible date formats and suggest different parsing options, showing you how well each option performs.

Dataiku also allows you to enrich data and create time-based features in just a few clicks, including:

• Subtract date components (month, hour, day of the week, week of the year, etc.)
• Calculate differences between date columns
• Marking national holidays
• ... and more!

‍

3. Clear complex text fields

When you have really messy text fields, let alone using complex regular expressions, cleaning and structuring that data is a nightmare. But Dataiku has code-free text clearing processors (splitting, finding and replacing, clipping, etc.) that make this possible in minutes.

Data from a JavaScript map on the web showing the name and location of garages:

‍

‍

And here, after just a few clicks, the same data is extracted as the columns containing the desired information:

‍

‍

4. Transferring Twitter Data

Dataiku has a simple connector that can retrieve Tweets or related information (user identifier, location, hashtags, etc.) based on keywords or hashtags. Once the data has been collected, visual text analysis features are similarable to cluster tweets, break them into words or n-grams, simplify and remove stationary words, and more.

5. Creating machine learning models

Who said you have to be a data scientist to use machine learning techniques? Dataiku allows non-coders to train algorithms, start making predictions, identify sets, and extract useful information about features without writing a single line of code.

‍

‍

6. Combining & joining datasets

Most of the time, data enrichment can be done by merging datasets — essentially we can define this as the process of getting columns from a dataset or tab to a reference dataset (VERTICAL). This is a key element of any analysis, but when you have multiple resources (both in terms of calculation time and aggregation criterion) this process can turn into a nightmare.

In Dataiku, the blending of data sources is simplified (and all visual, no VERTICAL) with specific processors that can easily retrieve all data from other datasets or combine them according to specific, fine-tuned keys and criteria.

7. Working with Geographic Data

Geospatial analysis is critical for a number of use cases.

For example, optimizing a network of rental agencies, mapping competition, sizing a target market, and more. Dataiku has several processors, which makes it easy to work with positions, in particular;

• Get latitude and longitude from an address with OpenStreetMap or Bing Maps API
• Enrichment of latitude and longitude with administrative information (city, state, department, etc.)

‍

The Dataiku graphics engine also provides the ability to draw scatter maps and heat maps with various levels of aggregation.

Discover the many things you can do without a single line of code in Dataiku right now and start experimenting.