According to Ricci. et al. 2011 recommendation engines, also referred to as recommender systems are applications or programs that uses various algorithms techniques to analyse, gather and forecast different kind of data in order to predict information and generate recommendations for users, more companies have recently started deploying recommendation engines in to their business’s (Ricci, Rokach, & Shapira, 2011). Amazon.com, YouTube, Netflix, Yahoo, TripAdvisor, Last.fm, and IMD, benefits from recommendation engines, that helps in growing their business and gets better understanding of the customer behavior online. For instance, Netflix, an online rental service films and movies, gave one-million-dollar award to a team that improved the performance of the company’s recommender systems filtering techniques and engine (Koren, Bell , & Volinsky, 2009). In other words, the Netflix competition example shows the growing importance of recommendation system for businesses online. Recommendation engines are valuable and useful software that handles and analyses massive amounts of data, then delivers a reduced and focused recommendations for users. The engines can be classified into different groups or categories, that each one of them represents and used for a specific purpose. According to (Jenson, 2017) The list of recommendation engines contains the following types : • (SaaS), known as software as a service recommender systemAccording to Amazon web services, Software as a service is an application delivery model which allows users to utilize a software solution over the Internet. SaaS engines are mostly payment based, so users must subscribe and pay a fee for the service. All types of recommendation system filtering techniques can be included in any (SaaS) application, in a recent study that was done by Afify et. al, 2016, proposed a (SaaS) recommender system and based it on a hybrid filtering that personalizes quality of service measurement, the proposed software as a service recommender system semantically manages user requests to detect business-oriented matching services, that are later ?ltered to fulfil the quality of services for user requirements and the service characteristics. In their work, the authors apply hybrid filtering techniques to authenticate the services set based on services metadata and the interests of the user. Lastly, the suggested set of services is arranged (M. Afify, F. Moawad, L. Badr, & F. Tolba, 2016). Software as a service is a growing model that includes extensive range of business, (SaaS) Recommender systems have many benefits such as: they mostly charge low fees, they are easy to integrate and to use, and they always provide various methods for improvement. However, (SaaS) recommender systems have some challenges in the development process, that involves dealing with multi-tenancy, repository and handling an enormous quantity of information and other softer concerns like keeping a client’s sensitive data safe on remote servers (Jenson, 2017).The following content contains examples for some of the (SaaS) recommender engines platforms that can be used for different domains and purposes such as data personalization and learning recommendations (Jenson, 2017) :a. http://retail.strands.com/ A platform for data personalization that uses (SaaS) recommender engine, the platform mostly focused on the product and e-commerce system, the platform offers real time recommendation, multiple personalization strategies and easily accessible from all devices such as laptops, cell phones and tablets. The platform’s recommendation engine has more than a hundred-customizable logic and personalization configurations that gives smarter filtered recommendations for user’s or the client’s customers. Figure 4.3: Strands Retail Product Recommendation Key Benefits (http://retail.strands.com/products/product-recommendations/, n.d.)The platform publishes lots of its clients’ case studies that used its recommendation engine and benefited from it. However, the platform doesn’t contain technical detailed information regarding how it’s recommendation engine is implemented and functioned in a client website. In order to setup the platform’s recommendation engine, the client must first upload the business catalog so that recommendation engine would learn the client’s products, after that it tracks the scripts and embed few tracking codes, so the engine would learn about the customers of the client and lastly it places recommendation widgets on the client’s website so that the customers can see the recommendations. b. Gravity R&DA Company that is established by the members of Netflix prize winners team “Gravity” in 2009, they created a platform know as (Yusp) that offers customized recommendations to users of websites, the company have many and well-known clients such as the DailyMotion and Cora Hypermarket. The platform mentions that the most important elements for it’s recommendation engine usage to generate predictions are the items, users, metadata and events. The recommendation engine analyzes the user behavior by embedding tracking codes on a website, which can be done with log on-site user activity that includes the user clicks, searches and item viewed, another way of analyzing user behavior is with an off-site user activity that includes Email tracing clicks and mobile applications. The platform (Yusp) uses both of collaborative and content based filtering algorithms techniques to automatically recommend relevant items and contents based on the data gathered from the user and moreover, it offers a unique dashboard which is used as a tool for dealing with all the recommendation engine associated settings on the user website.c. SuggestGrid A platform that offers a generic recommendation engine service that is full featured application programming interface and can be used in any area or domain in which user behavior and actions can be used for personalization. The platform explains briefly four steps for the process using its recommender engine, the following Figure 4.4 demonstrates the fours steps for using SuggestGrid recommender engine (https://www.suggestgrid.com/how, n.d.)..demonstrates the four steps for using SuggestGrid recommender engine (https://www.suggestgrid.com/how, n.d.)The platform recommendation engine is based on Apache spark which is a general engine for big data processing, but SuggestGrid has many improvements over it However, the platform is still in development according to the developers. The platform also contains documentation guide that explains in details different types of users’ actions including explicit and implicit, types of recommendation used and the similarities, metadata and its usage in solving cold start problem, different advanced features and how to learn from the previous errors.