A massive amount of data is generated globally in almost every area of society, including government, health care, business and research disciplines such as, biological sciences, natural sciences, engineering and social sciences. With inclusion of more and more data, we are also getting new insights and decision-making processes are also improving. Big Data can also be utilized to enhance quality of products and services. However, owing to the speed of generation of big data, we are still required to overcome the challenges related to efficient collection, segmentation, storage and processing of data. Applications of Big Data can be multidisciplinary, ranging from www.amazon.com  recommendation system to real-time surveillance of influenza outbreak .
Application of Big Data in Food Safety is a tedious task since the data and information related to Food Safety are spread across various sectors, including but not limited to agriculture, food and health sectors. To utilize big data for ensuring food safety, it is necessary to create and enforce standards that enable different systems to work together effectively and to ensure that sensitive information is protected.
What is Big Data
Big data is defined as extraordinarily massive and complex data collections that are difficult to process and analyze using typical approaches. It often originates from a variety of sources, such as social media, sensors or transactions and extracting valuable information from it requires the use of modern technology and analytical tools. Big data is significant because it can show patterns, trends and insights that would be difficult to uncover otherwise. It is used to make better judgements, anticipate future occurrences and increase efficiency in a variety of industries, including business, healthcare and research.
World Health Organization has adopted Ward and Berker’s definition  to Big Data which defines Big Data as “The emerging use of rapidly collected, complex data in such unprecedented quantities that terabytes (1012 bytes), petabytes (1015 bytes) or even zettabytes (1021bytes) of storage may be required.” To keep this in simple words, we can state Big data like a giant puzzle with millions of pieces. Each piece is a tiny bit of information, like a picture, a number or a word. Imagine, you want to see what the puzzle looks like when it’s finished, but you only have one piece. It doesn’t make much sense on its own. However, if you gather lots of puzzle pieces and put them together, you can start to see the bigger picture. Big data is similar. It’s a collection of lots of tiny bits of information that, when put together can help us learn new things, find patterns and solve problems. Just like you need to organize puzzle pieces and look at them closely to see the whole picture, we need special tools to help us to sort through big data and find the insights hiding inside.
Applications of Big Data in Food Safety
The WHO has lately adopted the big data approach to support food safety decision-making, resulting in the food safety platform “FOSCOLLAB” to allow integration of numerous sources from many disciplines. This platform houses various structured and un-structured data from multiple sectors and sources like the agriculture, public-health, food and animal sectors. These data indicators are embedded in several dedicated dashboard of the “FOSCOLLAB”.
The data sources linked in the dashboard are the databases on chemical risk assessment of the Joint FAO/WHO Meeting on Pesticide Residues (JMPR) and the Joint FAO/WHO Expert Committee on Food Additives (JECFA), the WHO database on Collaborating Centres and the GEMS food databases on food consumption and chemical occurrence in food. The “FOSCOLLAB” platform also shows possibilities for future expansions by foreshowing it’s ability to integrate furthermore data points from external data sources.
The following table shows some examples of data points included in “FOSCOLLAB” .
Big Data Workflow and its application in Food Safety
As we all know, managing big data is a tedious task. Thus, big data management is divided into various stages. These stages are depicted in the image below:
Data Collection Stage
There are many different places where we can find information that’s helpful for making sure our food is safe. These include online databases, the internet, omics profiling, mobile phones and social media. The tricky part is figuring out which information is important and how it’s connected to other sources of information. This is especially difficult when we’re dealing with sources that aren’t usually used for this purpose, like social media. In the next section, we’ll explore these various sources of data and talk about how they can be used to make our food even safer.
a) Online Databases
Online databases can be a significant stage in the collection of large data in food research. These databases can have a wealth of information about many parts of the food supply chain, such as food production, processing, distribution and consumption. Researchers and scientists may mine these datasets using big data analytics to find patterns, trends and insights that might assist in enhancing food safety and quality. Big data analytics, for example, may assist in identifying possible foodborne dangers, tracking the spread of foodborne diseases and optimizing food manufacturing operations.
The internet may be a great source of data for big data applications in food science. Researchers and scientists may acquire data on food safety and quality from a range of sources, including government agencies, academic institutions and business organisations, thanks to the large quantity of information available on the internet. Big data analytics can help researchers uncover trends and patterns in this data, allowing them to make better educated judgements about food safety and quality. Data from social media and online reviews, for example, can give insights into customer opinions and preferences, whilst data from food recall announcements can aid in the identification of possible dangers and guide regulatory choices.
c) Mobile Phones
Mobile phones have the potential to be a critical data gathering stage for big data applications in food science. With the increasing use of mobile phones, collecting data on all parts of the food supply chain, from food production and distribution to consumption and waste, has become easier than ever. Apps for mobile phones may collect data on a variety of variables related to food safety and quality, such as temperature, humidity and storage conditions. This data may be analyzed using big data analytics to find patterns and trends that can help influence food safety and quality choices.
d) Social Media
Social media can be a useful data collecting stage for big data applications in food science. Researchers and scientists may now access a great quantity of user-generated material linked to food, such as reviews, comments, images and videos, thanks to the widespread usage of social media platforms. Big data analytics may assist in the identification of patterns and trends in this data, offering insights into consumer behaviour and preferences, as well as possible dangers and risks linked with food items. Social media data, for example, may be used to detect development of any foodborne diseases, measure customer mood and satisfaction and guide marketing and advertising efforts.
Data Storing and Transferring
Traditionally, storing data has been done through the use of data management systems like PostgreSQL, Oracle and MySQL. However, when it comes to handling big data, these systems may not be enough. To efficiently manage large-scale data processing, faster, more flexible and reliable solutions are required. That’s where NoSQL databases come into play. These non-relational, open source and horizontally scalable databases have been developed as the next generation of data management systems. Popular NoSQL databases include HBase, Cassandra and MongoDB. But storage is only the beginning of the big data challenge. Once the data is stored, the next obstacle is transferring it from various sources into the NoSQL cluster for processing. This requires software that can handle large data transfers. Aspera and Talend are examples of such software used for moving big data.
The following table gives some examples used in the industry for data storage and processing :
After the storing and transferring step of data, it should be processed and analyzed. A huge list of analysis methods of data comes under the umbrella of Recommendation Systems and Machine Learning.
(i) Recommendation Systems
Recommendation systems are a type of big data application that uses machine learning algorithms to provide personalized suggestions to users. These systems analyse a user’s data, such as their past behaviour, preferences and feedback to recommend relevant products, services or content. As far as food safety is concerned, a recommendation system can be used to suggest safe and healthy food options to consumers.
For instance, a food safety application can collect data about the user’s dietary preferences, allergies and medical conditions. The system can then analyze this data and recommend foods that are safe for the user to consume. Additionally, the system can provide information about the nutritional content of different foods, including their calorie count, fat content and vitamin levels. Another example of a recommendation system in food safety is a platform that provides recommendations for food storage and handling. This system can collect data about a user’s kitchen habits, such as how they store food, how often they clean their refrigerator and how long they keep perishable items. The system can then provide personalized recommendations on how to store and handle food safely, such as how to keep meats separate from other items in the fridge and how to properly wash fruits and vegetables. Overall, recommendation systems are a powerful tool for enhancing food safety by providing personalized recommendations to consumers.
By using machine learning algorithms to analyze large volumes of data, these systems can help users make informed decisions about what to eat and how to handle their food safely. This system is used in e-commerce organizations to advice their buyers because of top selling products, history of any costumers buying. Basically, this system is made using data mining techniques like collaborative filtering, content-based filtering, etc.
(ii) Machine Learning
Machine learning is a type of data analysis technique that uses algorithms and statistical models to enable computers to learn from data and make predictions or decisions. In the context of big data, machine learning can be applied to analyze large volumes of data to discover patterns and insights that can be used to improve food safety. One example of a machine learning application in food safety is predicting foodborne illness outbreaks.
Machine learning models can be trained on historical data on foodborne illness outbreaks, including information on the type of pathogen, the food source and the geographic location. By analyzing this data, the model can identify patterns and risk factors that can be used to predict future outbreaks. This can enable public health officials to take preventive measures to reduce the risk of foodborne illnesses. Another example of machine learning in food safety is analysing sensor data from food processing plants. Many food processing plants use sensors to monitor temperature, humidity and other environmental factors that can impact food safety.
Machine learning algorithms can be used to analyze this sensor data in real-time, identifying potential hazards or anomalies that could indicate a food safety issue. This can help food processing plants take preventive measures to reduce the risk of contamination and ensure that their products are safe for consumption. Overall, machine learning is a powerful tool for analyzing big data and improving food safety. By analyzing large volumes of data, machine learning algorithms can identify patterns and insights that can be used to predict and prevent food safety issues. This can help to reduce the risk of foodborne illnesses and improve the safety and quality of our food supply.
Several visualization methods are available to analyze and present summarizes of huge amount of data in simple ways like pie-charts or bar-charts. Some examples of such software are R, Cytoscope etc..
Practical Application of Big Data in Food Safety
Agricultural chain and food supply chain – In this sector, for detecting pathogens and contaminations big data can be used by linking the information on various environmental factors. Several data collection and analysis systems have been developed for supporting farmers in decision making, etc. By combining massive datasets and integrating it with algorithms and tools, it can be very useful in terms of uprising agriculture industry. One example for this could be www.semagrow.eu. This system queries large number of datasets and uses various tools to help in decision making. Another example would be www.trees4future.eu with the help of this project, the enormous amount of forestry scientific data accessible to the public and scientific community.
In conclusion, big data is playing an increasingly important role in food safety, as it can provide new insights and improve decision-making processes. However, its application in food safety is challenging, as the data is spread across various sectors. Thus, it is necessary to create and enforce standards that enable different systems to work together effectively and protect sensitive information.
The Food Safety platform “FOSCOLLAB” is an example of how big data can be utilized to support food safety decision-making. Future expansion of the platform can integrate data points from external data sources. The management of big data is divided into various stages that include data collection, processing, and analysis and each stage is equally important. The potential of big data in food safety is vast and has a lot of scope for the future. With technological advancements and increased collaboration between different sectors, big data can transform food safety and reduce the risk of foodborne diseases.
1. Linden, G., Smith, B. and York, J. (2003b). Amazon.com recommendations: Item-to-item collaborative filtering. IEEE Internet Comput. 7:76–80.
2. Ginsberg, J., Mohebbi, M. H., Patel, R. S., Brammer, L., Smolinski, M. S. and Brilliant, L. (2009). Detecting influenza epidemics using search engine query data. Nature. 457:1012–1014.
3. Ward, J. S. and Barker, A. (2013). Undefined by data: A survey of big data definitions. CoRR. abs/1309.5821.
4. Hans J. P. Marvin, Esmée M. Janssen, Yamine Bouzembrak, Peter J. M. Hendriksen & Martijn Staats (2017) Big data in food safety: An overview, Critical Reviews in Food Science and Nutrition, 57:11, 2286-2295, DOI: 10.1080/10408398.2016.1257481
About the Authors:
The views/opinions expressed by authors on this website solely reflect the author(s) and do not necessarily reflect the views/opinions of the Editors/Publisher. Neither the Editors nor the Publisher can be held responsible and liable for consequences that may arise on account of errors/omissions appearing in the Articles/Opinions.