DEV Community

Cover image for CHALLENGES FACED IN THE DATA SCIENCE IN TODAY’S MODERN ERA
Tejaswiniteju
Tejaswiniteju

Posted on • Updated on

CHALLENGES FACED IN THE DATA SCIENCE IN TODAY’S MODERN ERA

A recent survey of 16,000 data enthusiasts, concluded that the most occurring and frequent challenges faced in the data science are nothing else than the corrupt data (approx 36%), unavailability of talent in masters in data science (about 30%) and no management support (27%) at all. To study this issue, I have preferred data from Kaggle collected during August 2017.

BARRIER AND HINDRANCES CREATED AT WORK
It has been asked in the survey by the respondents that what are the major problems faced by you all in the past years. Results appear and the most important 10 challenges were as follows:
Corrupt Data (36%)
Unavailability of talent in data science (30%)
Politics in the companies (27%)
Unclear questions (22%)
Can’t access data (22%)
Privacy problems (18%)
Inexperienced experts in the domain (16%)
Small organization and unable to afford a data science team (14%)
Explanation of data science in front of others (14%)
Results were not used by during the decision making process (13%)
After going through the results, it has been concluded that on an average a data scientist is facing three (median) barriers in previous years. The challenges faced varied on the basis of the job profile a person is having in the field of data science. Those who are programmer reported one and those who are data scientist reported 4 platforms.
PLATFORMS GROUPED TOGETHER BY PCA
A survey conducted in which the Principal Component Analysis has been calculated for 20 challenges in order to group the considered challenges.
The top 5 components are as follows:
RESULTS NOT UTILISED IN THE DECISION PROCESS:
These types of challenges come under the politics which occurs in a company, unavailability of the management support and inability to utilize the findings earned during the decision making.
DATA PRIVACY AND UNAVAILABILITY:
These types of challenges dealt with the data only that how much clean data is available, what are the privacy aspects of the data and the availability of the data.
UNAVAILABILITY OF FUNDS:
Challenge under the lack of funds relates to the organization that how much the organization buy from the out sources, domain expertise of the organization, and the data science talent.
WRONG QUESTIONS ASKED:
Not having clearance in the question to be asked and to be answered in the data science field in order to go in a clear direction to the available data.
LIMITATIONS OF THE TOOLS USED:
These types of challenges include the tools which are used for the extraction process, deployment of the available models as well as for the scaling of the database to a higher level.

Data scientists are the bridge between the IT sector and the Management department. Hence, there is a need to manage these challenges for future development.

Resource Box:
If a person really wants to become a good Data Scientist then he/she needs to go through both the good as well as the bad aspects of the field and try to get rid of these challenges faced. data science course malaysia institute is one of the best institutes for the data science course. Hence, apply over it.

Click here to know more about data science course

Address: 360DigiTMG - Data Science, IR 4.0, AI, Machine Learning Training in Malaysia
Level 16, 1 Sentral,, Jalan Stesen Sentral 5,, KL Sentral,KL Sentral50470 Kuala Lumpur, Malaysia
phone no: 011-3799 1378
Youtube: https://www.youtube.com/watch?v=21MSMs34bSc

Discussion (1)