The challenges of implementing data science

What is data science?

Data science is an interdisciplinary field that uses methods, processes, algorithms and scientific systems to obtain value from data. Data experts combine a wide range of skills - including statistics, computing and business knowledge - to analyze data collected from the web, smartphones, customers, sensors and other sources.


Data science reveals trends and produces information that businesses can use to make better decisions and create more innovative products and services. Data is the foundation of innovation, but their value comes from the information that data experts can gather and can then act upon.

The challenges of implementing data science

1. Despite the promise of data science and the huge investments in data science teams, many companies do not realize the full value of data. In their pursuit of hiring talent and creating data science programs, some companies have experienced inefficient team workflows, with people using different tools and processes that do not work well together. Without a more disciplined and centralized management, decision-making roles may not achieve a complete return on investment. This chaotic environment presents many challenges.
2. Data experts cannot work effectively. Because access to data must be granted by an IT administrator, data experts often wait a long time for the data and resources they need for analysis. After gaining access, the data science team could analyze the data using different and possibly incompatible tools. For example, a scientist might develop a model using the R language, but the application in which it will be used is written in another language. For this reason, the implementation of models in useful applications can take weeks or even months.
3. Application developers cannot access usable machine learning. Sometimes the machine learning models that developers receive must be recoded or not ready to be implemented in applications. And because access points can be inflexible, models cannot be implemented in all scenarios, and scalability is transferred to the application developer.
4. IT administrators spend too much time providing support. Due to the proliferation of open source tools, the IT sector needs to provide support for a growing list of tools. For example, a marketing data expert should use tools different from those used by a finance data expert. Teams could also have different workflows, which means that the IT department should constantly rebuild and update environments.
5. Business managers are too far from data science. Data science workflows are not always integrated into business decision-making processes and systems, which makes it difficult for business managers to collaborate with data experts. Without better integration, business managers cannot understand why it takes so long to move from prototype to production - and are less likely to support investments in projects they consider too slow.

This article was originally published on   -------------------------More info