Data Analytics

Data Mining in Agriculture {0}

By: Paola A.

Every year, technology changes and new developments help many economic sectors discover new ways to improve, forecast a change, etc. For example, one economic sector that is benefitting from using new developments in technology is agriculture. In agriculture, a way to discover this type of changes is through data mining. But what exactly is data mining and how is agriculture benefitting from this. “Data mining involves the process of finding large quantity of previously unknown data, and then their use in important business decision making” (Milovic & Radojevic, 2015). For instance, in developing countries such as India, using data mining for “price prediction helps the farmers and also Government to make effective decision[s]” (Hemageetha & Nasira, 2012). Furthermore, data mining is something that could also help consumers by preparing them in case of any change in prices. Finally, this method of collecting data is something that can definitely improve and benefit the way farmers, government, and consumers make better decisions in the future with different applications, but it also has challenges and limitations.
Using data mining in agriculture benefits farmers, government, etc. in many ways. One of the ways that using data mining is beneficial in agriculture is “possibility to study hidden patterns in datasets in agricultural domain. These patterns can be used for diagnosing crop condition, prognosing market development, monitoring customer solvency” (Milovic & Radojevic, 2015). In other words, it can help farmers tell around the time that their crop would flourish, make predictions of when their product is more likely to sell, and what product customers are buying more. Also, another benefit is that “Agricultural institutions use data mining technique and applications for different areas, for instance agronomists use patterns measuring growth indicators of plants, crop quality indicators, success of taken agro technical measures and managers of agricultural organizations pay attention on user satisfaction and economically optimal decisions” (Milovic & Radojevic, 2015). read more...

Data Mining in Marketing {0}

By: Mohammed A.

Have you ever experienced this, where you open your social media page and all of a sudden see advertisements of a shirt or pair of shoes you were looking at a day ago; well know that it is not a coincidence. It is very likely that the company has been monitoring your online habits for some time and used that data to market and advertise specific products that align with your interests. Within the journal article “Revisiting the problem of market segmentation,” the author emphasizes how data mining helps marketing users to target marketing campaigns and also to align campaigns with the needs, wants, and attitudes of customers and prospects (Lien). This process is most widely used digitally, where every website a person visits leaves a digital footprint, which allows companies to gather that data and market products that will appeal to the individual the most. In essence, data mining can be very beneficial in the advertisement and market industry, where companies are able to gain a competitive advantage through the use of various tools and techniques that allow them to monitor their potential customers’ behaviors. read more...

Predictive Analytics {Comments Off on Predictive Analytics}

By Andrew C.

Predictive analytics, a technique of using data metrics, current and historical, as well as predictors to determine a future outcome. It is also referred to as predictive modeling or forecasting. In essence, one could say it is an attempt at predicting the future, while a bit simple it is not entirely incorrect. Although it sounds outlandish, predictive analytics is very much a core aspect of our lives today, and it is very real.
The benefits of predictive analytics are obvious if you can imagine what kind of advantages predicting future outcomes might give you. The overall premise is that it can drive competitive advantage through a number of ways: identifying trends, understanding customers better, aiding in strategic decision making, predicting industry behavior, process and performance optimization among others. Generally speaking, it can provide a more accurate picture of your business environment; your customers, operations, threats, opportunities, etc. In point of fact, Harrah’s Entertainment, a casino conglomerate uses predictive analytics to drive its marketing and operational decision making and in 2003 it saw an increase in income from operations of 26.6% (Felipe-Barkin, 2011).
Predictive analytics is used in many industries, though there is heavy use in financial services, insurance, retail, telecommunications, healthcare, and pharmaceuticals. In fact, predictive analytics is an aspect that affects our lives every day – credit scoring. Credit scores, an estimate on your ability to pay your bills, maintain debt, etc. in the future are driven by predictive analytics statistical mode and are an everyday facet of our lives today (Brown, 2015). Not only business aspects are affected by predictive analytics. In the early 2000s Billy Beane, the general manager for the Oakland Athletics, with the help of Paul DePodesta used predictive analytics and modeling to put together a team that could still compete with its richer competitors – and they went all the way to the playoffs. A remarkable feat all things considered.
While predictive analytics is a powerful tool, it should be noted that it is simply a tool. It is not a crystal ball or an oracle. It requires large amounts of data, both historical and current, as well as the tools and personnel to organize and interpret that data and turn it into meaningful information. read more...

Analytics Tools: Flurry vs. Mixpanel {Comments Off on Analytics Tools: Flurry vs. Mixpanel}

By Caleb W.

For mobile app developers, knowing how their target audience uses their product is an important part of the app’s development cycle. Knowing demographic information about the user, like their age and gender, help developers keep the app relevant to their interests. Tracking usage of the app allows developers to see where users are having difficulty with the app interface and streamline it for a more pleasant user experience. Developers can focus their efforts on improving key functions by logging user activities, like creating accounts, posting comments, and liking content. Integration of analytical tools into apps allow for a wealth of information into how apps are used and the people who use them.
Implementing analytical tools is a typically straightforward process, although ease of use can vary among different software. After importing the software library, some tools may require additional calls within the code in order to track events. The software is able to associate all instances of an app through a uniquely generated API token. Once conditions are set, the app will report when the event defined by the corresponding condition occurs. The developer can then segment the data, filtering it by a specific demographic. This gives developers the information they need to cater to the specific needs of each of their users.
Flurry is an effective analytics tool that can be deployed easily and quickly. Many of its users noted its support for many different platforms and the ease of integration of its SDK. Flurry also has built-in installation tracking, allowing the developer to see the ads that resulted in the user downloading the app. Flurry is free to developers as well, and also has the ability to push ads and notifications to the user. Being free software, however, comes with several drawbacks. Flurry is unable to process data in real-time, usually taking about a day to generate reports. It has limited setting of conditions and segmentation, thus depriving developers the chance to dig deep into certain parameters they may want to track. This inflexible reporting is compounded by several user complaints that some metrics needed to be defined more accurately. Flurry also lacks detailed external support for its API, limiting the methods by which apps can report usage data. Nevertheless, for a free product, Flurry is perfect for developers who simply want to get the ball rolling on mobile analytics.
One of Flurry’s more robust competitors is Mixpanel. Although developers reported difficulty in understanding and integrating the library, they also widely complemented the ease of usage of the app’s interface and its ability to optimize the end user’s experience. The app also provided concise reporting of data and allowed developers to easily set conditions and segment the funneled data. Along with reporting data within seconds of the corresponding events being triggered, Mixpanel also provided support for third-party installation tracking services, as well as a publicly documented API. Mixpanel’s biggest disadvantage, unsurprisingly, was its cost, with its only free option limiting event reports to 25,000 data points or 1,000 unique profiles. Given all the benefits that come with it, however, Mixpanel makes a much better choice for developers needing much more breadth and depth in usage analysis.
Tracking mobile app usage in today’s technological age is a must for any developer. With app stores filled with so many different apps performing so many similar functions, it is important for an app to stand out to users in its functionality and ease of use. For the very low price of free, Flurry is able to provide basic analysis of user activity and demographics. If developers want more insight and control into their data, however, Mixpanel fits the bill. Failure to take advantage of mobile analytics software, however, leaves developers in the dark on how to effectively cater to the needs of their app’s clients.

Works Cited
Flurry Analytics. Retrieved from https://developer.yahoo.com/flurry/docs/analytics/
Help Center. Retrieved from https://mixpanel.com/help/
Lin, Y. (2014, July 30). App Analytics Strategies & Tools. Retrieved from http://blog.kiip.me/developers/mobile-app-analytics/
Khorkov, E. (2015, March 6). Mobile analytics: Mixpanel vs Amplitude vs Flurry vs Localytics. Retrieved from https://medium.com/polecat-blog/mobile-analytics-mixpanel-vs-amplitud-vs-flurry-vs-localytics-aeb6bf02b734#.2mnklr62f read more...

Analytics for Online Learning {Comments Off on Analytics for Online Learning}

By Chad C.

One of the main definitions for analytics is the “measurement, collection, analysis, and reporting of data about learners and their contexts, for the purposes of understanding and optimizing learning and the environments in which it occurs” (Hampson). It is a fact that technologies are being constantly upgraded, and previous forms of instruction are not as suitable to today’s students. Because of this, Massive Open Online Classes (MOOCs) have been created, and online classes are now more readily available. This means that students are able to complete courses fully online, rather than having to be physically present and face to face with an instructor. Tuition is steadily rising and students are becoming more and more busy with work, internships, and other obligations that might not necessarily allow enough time for them to go to university campuses and sit though hours worth of lectures. The availability and convenience of online classes offers students the classes that they need, as well as the flexibility in schedule that they desire. Big data analytics for online learning provides benefits in terms of bettering retention and completion rates, tracking at-risk students, and aligning institutional aids with participant satisfaction. read more...

Analytics for Online Learning {Comments Off on Analytics for Online Learning}

By Bryan V.

Remember the time where the only place you could learn educational material was through a teacher or from a book at your local library? Our technological skills and equipment have increased tenfold since then and with that has come Learning Analytics. Learning Analytics is the process of collecting and analyzing data for the purpose of improving online learning environments. Our technology has become so advanced that you can learn almost anything from the comfort of your own home. Learning analytics has helped create most of the learning tools available through the internet today. In this post I will talk about the benefits, tools/software used in Learning Analytics, applications of Learning Analytics, and some limitations concerning the use of Learning Analytics. read more...

Competitive Advantage through Recommender Systems {Comments Off on Competitive Advantage through Recommender Systems}

By Chundyanto W.

It is hard to imagine the existence of e-commerce websites or any other popular entertainment and social media sites that do not utilize recommender systems nowadays. Recommender system is a technology to filter information in a website or a system in order to predict the rating or preference of a product for users. It has made a massive change in the way people interact with websites. In e-commerce websites, for example, users are easily guided to products they like according to their preference and taste based on their past shopping information. Another example can be taken from social networking sites, where users are suggested to connect with other users they may know based on their mutual interest, friends, or occupation. This improvement has certainly created a more attractive online experience for users by giving an ease of access through recommendations from the system.
Aside from providing an ease of access to guide users in interacting with a website, there are some other important benefits that are produced by the recommender systems. Recommender systems deliver relevant content to users because users will only be directed to items that match the users’ preference. This will lead to another benefit, which is assisting users in personalizing their profiles within the website. These efforts are meant to make users feel comfortable with the website, so that they will turn from just “regular shoppers, into loyal customers.” (Certona) Thus, not only it will increase customer satisfaction, but it will also help online businesses generate more profits from retaining customers.
Recommender systems can be classified into three categories. They are “Collaboration Based Recommender System,” “Content-Based Recommender System,” and “Hybrid Recommender System.” (Jones) First, collaborative filtering is constructed from users’ behaviors with similar traits. For example, users that buy book A are likely to buy book B, therefore other users that buy book A will be recommended with book B. Next, content-based filtering predicts users’ preference based on “the basis of a user’s behavior.” (Jones) Users that buy books about analytics, comment, and subscribe to analytics newsletter, will be suggested products or news that relate to analytics. Lastly, the hybrid approach is a combination of collaboration filtering and content-based filtering. This combination of approaches enhances the performance and usability of recommender systems in websites.
Some tools that functionalize recommender systems include LensKit, Scout Portal Toolkit (SPT), Recommender.org, and Duine Toolkit. To learn about how the tool works, we can take LensKit, a free open-source software to build recommender systems, as an example. LensKit allows us to create recommender algorithms through two main utilities, which are the “ItemScorer” and “RatingPredictor”. Based on the user’s profile, trait, and behavior, any numbers can be put into ItemScorer, then evaluated by RatingPredictor as an algorithm to determine whether the score matches or is similar to a related item. ItemScorer’s number can be based on purchase probabilities, which will then be scored by RatingPredictor to output the predicted ratings to determine the user’s preference and relation with a certain item. (LensKit) In addition, an interface called “Top-N recommendation” can organize recommended items that are scored in ItemScorer for users based on their user ID. (Lenskit) This way, an algorithm can be implemented into a website or a system to utilize recommender systems.
Big names such as Amazon, Ebay, Facebook, YouTube, and Netflix adopt recommender systems to be successful in their performance. To increase performance efficiency, most of these websites utilize the hybrid approach, or more than one recommender systems approach in their system. For example, Amazon uses the collaborative filtering to determine the likelihood of a customer with a certain trait to buy a product, as well as content-based filtering to narrow the product’s criteria to match the user’s preference. Netflix, the video rental and streaming service giant, flawlessly utilize the hybrid approach as well. It uses the collaborative filtering to see if users that watch a certain genre tend to watch the other genres or related movies. Collaborative filtering uses this information to make “individualized predictions.” (Bell, Koren, and Volinsky) Then it uses content-based filtering to narrow down the list of movies that may interest the user.
Tremendous functionality comes with a price, which applies to recommender systems. Although recommender systems work really well in terms of assisting users, there are challenges and limitations for the systems. First of all, a sufficient amount of data is needed to help predict users’ preference. It is hard for the system to predict a new user’s preference if the user only provides a limited amount of information and profile, especially if the user does not show a consistent pattern of behavior in the system. Changing user’s preferences can potentially be a problem as well. If the system only retains users’ past information, it will not recommend users based on their new preference. This will lead to a decrease in users’ satisfaction. (Muthukumar) Lastly, recommender systems are complex and can be difficult to implement.
It is clear that recommender systems open up a new challenge for online sites and businesses to compete for a competitive advantage in retaining their customers. Users always seek online experiences and transactions with websites that match their preference, which is mainly manipulated by recommendations from the system. Hence, websites with excellent recommender systems are likely to be successful.
Keywords: recommender systems, competitive advantage, benefits, limitations read more...

Analytics for Content Websites {Comments Off on Analytics for Content Websites}

By Jeffrey T.

Content websites, as their name implies, are sites dedicated to featuring content or information like news, articles, or blog posts whether it be in the form of text or multimedia. Given the diverse range of options for the types of content that can be posted, there is also a wide variety of analytics that can be employed to help maximize the visitor turnout, time spent on a page, and general web traffic. The purpose of utilizing analytics for content websites is to find out more about your users and the efficacy of your website in terms of its layout, marketing methods, and a whole slew of other categories in order to optimize and popularize your content.
For any website owner, using every form of analytics would yield the greatest returns whether they gather descriptive, diagnostic, predictive, or prescriptive analytics; however not all of them have the time nor the money to invest in them. Thankfully, there are free web analytics like the universally recognized Google Analytics, which is often more than enough information for most people or businesses. With the American Cancer Society, a 5.4% increase in donations resulted when they used Google Analytics to segment their traffic into three categories and cater to each one based on what they were seeking. Donors, event participants, and information seekers were recognized and tracked for their page views, substantial donations, or events completed through Search Discovery and Custom Dimensions: premiums of Google Analytics. Through analysis, they also discovered a large portion of their user base was mistakenly visiting their site during Breast Cancer Awareness Month, and, in response, they successfully redirected them to their Making Strides website with new promotions and links.
Focusing on their In-Page analytics feature, I found that Google Analytics successfully employs click analytics to help users identify and partition their viewers into groups. Every click tells us a little more about someone based on where they navigate to and the time they spend on certain pages. Other click analytics tools such as ClickHeat concentrate on heatmapping user clicks, which paint a more visually simplified picture of where users are clicking in a website using warm colors. The Kraemer Family Library site owner found that users constantly clicked on graphics near related text only to find out that they weren’t links, and made the appropriate corrections. Unfortunately, this type of mapping does not emphasize the numbers of clicks or give much information about who is clicking. By itself, ClickHeat’s use is limited, so it’s often paired with Piwik, another web analytics tool, as a plug-in.
If you decide to use these tools, you’ll still need to spend a substantial amount of time learning, and then using these applications for data analysis. There may be great flexibility within these programs, but nonetheless, it will be necessary to continuously spend time examining information collected, making the corresponding changes, and then viewing the results of those changes in a repeating methodical cycle to keep things up to date. Basically, the increased success of your content website will often prompt greater time consumption or resource allocation. Another issue could be the misunderstanding or misinterpretation of data. Analytics are fantastic tools for gathering information, but are better treated as supplements than complements, because it may be more necessary to learn the fundamentals of business and data first to properly use the analytics. A lack of understanding in those fields could lead to misusage of data and influence you to make suboptimal decisions or misinterpret trends as well as the meaning behind your data. This is not to say that these tools are flawed in that respect, but that we ourselves need to be proficient enough to avoid the common pitfalls in the vast library of information.
Overall, there is a great deal to learn with analytical tools and software; however, we must prepare ourselves just as thoroughly as we prepare the content on our website; otherwise we risk converting our assets into liabilities. read more...

Content Analytics with Content Websites {Comments Off on Content Analytics with Content Websites}

By Don T.

In today’s day and age, the playing field is constantly changing, whether it is the tastes of the consumers, or the trends that follow from a certain event or thought. In order to keep up with these changes, in order for companies to remain competitive in today’s world, and for other groups to keep and gain others’ interest for their group or activity, they will most likely do business with companies that utilize content analytics. By using content analytics, these businesses, groups, and individuals will be able to analyze and change as quickly as these trends do and to be able to keep and gain more audience members.
Content Analytics is a term used to refer to a variety of operations, including web analytics, content assessment, metadata tagging, social media monitoring, sentiment analysis, and text analytics (Seymour 2013). In short, content analytics refers to the analysis of the content presented on a web page, the data that is involved with it, what a user does whenever the data displayed on the page is accessed or touched upon by the user, and how the user reacts to the content that is on the page. The page is also analyzed with what users read the most, what users used to find the web page in question, what keywords were most important into drawing users to the web page, and what other content they searched for using that same page. It is the culmination of using many different types of analytics like Web Analytics, Visual Analytics, Social Analytics, Descriptive Analytics, and many more.
The benefits to using content analytics are that it will greatly improve a company’s ability to see how well their content is displayed, what solutions can be implemented should there be a problem with their display and their content, and that it can draw in more business and interested people. An example might include how an amusement park, like Disneyland, is able to set up their web pages in order to draw in those of whom are interested in either visiting or researching more information about the parks, hotels, restaurants, shops, and other venues they might like to come visit. If their page is set up in a way that works well with their users, it can draw in more people to visit their parks because of how well their information is displayed, how much information that the people using it receive, and how well the user is able to navigate about the website using their specified interests, such as “Thrill Rides”, “Fireworks show”, or “Character Greetings”. If the users are able navigate the pages easily and are able to use ideal keywords to bring them to the pages that they are looking for, their readers will be more influenced to make a visit out to the parks or surrounding hotels, shopping, and eating venues. The same could be done with news websites, such as FOX News or CNN News. FOX and CNN display news articles on their website, in which they use content on their page to tell the stories and specific keywords in order to draw in their desired readers. Once those readers have accessed the page in question, links to other articles that contain related keywords will begin to appear, in which case readers will be able to choose where they want to go if they wish to continue reading articles from that source. Using content analytics with these sorts of websites will help to greatly improve the quality of the content displayed and to help increase traffic to those pages.
Some examples of software programs that do content analysis are Google Analytics, IBM Watson, and Content Analytics. Google Analytics does more types of analyzations, such as web analytics, but web analytics is a tool that can aid in the analyzation process of content analytics. For example, if a website that has used web analyzation and visual analyzation is not doing so well with traffic, thanks to how the website is set up and because their visitors aren’t really reading the content on the page, the creators of the website could consider changing the layout and content of the website in order to draw in more visitors and increase its popularity. The company Content Analytics specifically uses “Enterprise level analytics tools for optimizing product content, keyword relevance & rankings in global e-commerce” (Content Analytics 2016). Content Analytics aids companies in using their analytics tools for helping customers with changing their content in order to draw in more customers and visitors.
The limitations of using content analytics for websites is that it is incredibly difficult to do. There are not many companies out there today that do website content analytics in today’s world. Another limitation is that it is extremely time consuming. Although there are companies that have produced software to accomplish this task in a shorter amount of time, it still takes a considerable amount of time to do. Web content analytics can also mislead companies about trends that are currently happening with their content, such as an article being something talked about in high volume like presidential candidates and their thoughts and not towards something that is less talked about in society, such as a localized natural disaster, like a flood or an earthquake. Another limitation is that web content analytics “… cannot tell us what people really think about these images or whether they affect people’s behavior” (Crossman 2016). This means that although companies can analyze the content, they will not be able to tell what the users and readers think about the content being published; they can only see whether it is being accessed in a high volume based on keywords or images sending them to the page with the content containing those keywords or images.
Although the method of content analytics has been around for many years, web content analytics is something that is highly sought after and will be for many years to come. In order to be able to survive in today’s market and internet geared society, groups, organizations, and companies will use content analytics in order to evolve their website content towards their desired readers, users, and customers. read more...

Analytics for SaaS {Comments Off on Analytics for SaaS}

By Clanesha S.

One of the greatest innovations that has occurred in technology is being able to provide software applications over the internet. Today, businesses no longer find it necessary to physically purchase and install applications. Instead they are able to access applications by simply being connected to the internet. This breakthrough in technology is known as “Software as a Service” (SaaS). So how can businesses that incorporate SaaS know how grow their company? How can they come up with new ideas to increase revenue without having the burden of installing and managing software tools?
In order to answer these questions it is important to understand SaaS analytics and its importance. SaaS analytics is a web based software that collects data from organizations in order to improve their business. Analytics for SaaS based companies are usually based on monthly subscriptions meaning that a company can cancel their subscription at any given time. Some examples of SaaS based companies are (Google Apples, Salesforce, and Dropbox). In addition, SaaS based companies have immediate access to analytics which helps companies plan for a successful future.
It is important to mention that in order for SaaS based companies to survive today it is necessary to incorporate SaaS analytics. According to research published by Forrester the reason for the growth in SaaS is due to companies being driven by customer demands. In order for a SaaS based companies to know their next move it is important for them to focus on “what the customer is doing and what the customer wants.”
So how does a company incorporate SaaS analytics? There are many analytics SaaS providers such as Woopra, IBM, and GoodData. These providers’ helps companies understand and analyze trends, improve products, and provide the best solutions for success. While some analytic SaaS providers provide better features than others. It is important for SaaS based companies to determine which provider would be the best choice for their company. When a company receives data from analytic providers they are able to see key performance indicators such as subscription changes and know exactly how they can improve. A company can even get ahead of the customer by analyzing retention reports which shows how long until a customer cancels, upgrades, or downgrades their subscription. So a SaaS company that relies heavily on customers subscriptions can benefit from retention reports.
Also, there are many other benefits when a company incorporates SaaS analytics. Some of these benefits includes reduced cost, improve customer experience, instant access, and overall improvement of the company. In addition, with SaaS Analytics a company can determine how to increase revenue by choosing the right key performance indicators. In order for a company to choose the right KPI’s they must choose wisely and avoid analyzing to many KPI’s at once.
Tools that a SaaS based company might use for SaaS Analytics would be (Birst, GoodData, Tableau, or Google Analytics). While each of these tools offer different features they all share common characteristics. Each tool is able to track the movement of a customer, report revenue changes, offer many key performance indicators to choose from, and help companies improve their service. While there are many benefits of Analytics for SaaS it is important to point out that SaaS based analytics does possess a few challenges and limitations. Challenges include integration, security, performance, and functionality. Integration is a major issue SaaS analytics due analytics SaaS providers applications do not work with customers existing applications. In addition, some SaaS analytic providers have tried to come up with their own customary integration methods however the cost was expensive and the design was difficult. Another major problem is security. Yes! Security is a huge concern due to companies having confidential information stored on a cloud and former employees having access to the company’s analytics. For companies to restrict access it becomes a long process which some companies may fail to do. read more...