1. Personalized Recommendations

One of the most notable applications of machine learning in e-commerce is personalized recommendations. By analyzing user behavior, purchase history, and browsing patterns, ML algorithms can suggest products that are tailored to individual preferences. This not only enhances the shopping experience but also increases the likelihood of conversion.

Case Study: Amazon’s recommendation engine, which contributes significantly to its sales by suggesting relevant products to users based on their browsing and purchasing history.

2. Dynamic Pricing

Dynamic pricing involves adjusting the prices of products in real-time based on demand, supply, competition, and other factors. Machine learning models analyze vast amounts of data to determine the optimal price point, maximizing profits while remaining competitive.

Example:Airlines and hotel booking sites often use dynamic pricing to offer competitive rates that reflect current market conditions.

3. Inventory Management

Efficient inventory management is crucial for e-commerce businesses. Machine learning can predict demand trends, helping retailers maintain optimal inventory levels. This reduces the risk of overstocking or stockouts, ensuring that popular items are always available while minimizing storage costs.

Case Study: Zara uses ML to predict fashion trends and manage inventory, ensuring that its stores are stocked with the latest trends.

4. Customer Segmentation

Machine learning enables e-commerce platforms to segment their customer base more effectively. By analyzing data such as purchase history, demographics, and browsing behavior, ML algorithms can identify distinct customer segments and tailor marketing strategies accordingly.

Example: Targeted email campaigns that cater to different customer segments, increasing engagement and conversion rates.

5. Fraud Detection and Prevention

E-commerce platforms are vulnerable to fraudulent activities. Machine learning models can detect unusual patterns and behaviors indicative of fraud, such as unusual purchase volumes or atypical login locations. This allows businesses to take proactive measures to prevent fraud and protect their customers.

Example: PayPal uses ML algorithms to monitor transactions in real-time and flag suspicious activities, reducing the incidence of fraud.

6. Customer Service Automation

Machine learning-powered chatbots and virtual assistants are becoming increasingly common in e-commerce. These tools can handle a wide range of customer inquiries, from order status to product information, providing quick and accurate responses while freeing up human agents for more complex issues.

Case Study: Sephora’s virtual assistant, which helps customers find products, offers makeup tutorials, and provides personalized beauty advice.

7. Image Recognition

Image recognition technology, powered by machine learning, allows customers to search for products using images. This is particularly useful in fashion and home decor, where customers may want to find items that match a particular style or design they have seen elsewhere.

Example: ASOS’s visual search feature, which lets users upload photos to find similar products on the platform.

8. Enhanced Search Functionality

Machine learning improves the accuracy and relevance of search results on e-commerce platforms. By understanding user intent and context, ML algorithms can deliver more accurate search results, enhancing the overall shopping experience.

Case Study: eBay’s use of natural language processing (NLP) to improve search results and help users find exactly what they’re looking for.

Conclusion

Machine learning is transforming the e-commerce landscape, offering numerous benefits from personalized shopping experiences to efficient inventory management and robust fraud detection. As ML technology continues to evolve, its impact on e-commerce will only grow, paving the way for more innovations and improved customer experiences. Embracing machine learning can provide e-commerce businesses with a significant competitive edge, driving growth and customer satisfaction.

Artificial intelligence (AI) has become an non-separable part of our lives and its impact can be seen across all industries, including marketing and advertising. The use of artificial intelligence in these areas has increased over the past few years, and the benefits are becoming more apparent. From automating tasks to delivering personalized experiences to customers, AI is proving to be gamechanger in the world of marketing and advertising.

What Is Artificial Intelligence in Marketing?

AI Marketing leverages user data and artificial intelligence concepts such as machine learning to predict the customer’s next step and improve the customer journey.
The chart below shows how marketers are integrating AI and machine learning into every step of the customer’s life.

If you haven’t thought about the power of AI in business yet, now is the time to learn more. We’ve compiled some great AI marketing examples to get you started.

1. Amazon — When shopping on Amazon, you may see options such as “recommended for you”, “Items you liked” or “Customers also bought”. This is one of the most popular AI-powered strategies Amazon uses to drive sales, powered by Amazon’s powerful recommendation engine. Amazon leverages product-to-product integration to power its recommendation engine. Unlike content-based filtering, collaborative filtering uses other users’ information to generate recommendations.
2. ClickUp Increases Blog Traffic by 85% Using NLP AI — Project Management ClickUp Uses Surfer SEO’s Natural Language Processing Artificial Intelligence Tools and Machine Learning To:
   • Identify content optimization methods;
   • Understand what elements to include in the product (and density);
   • Get an idea of ​​the best article structure, from the number of images included to the length of subheadings.
3. Starbucks uses predictive analytics to deliver personalized recommendations — Starbucks is an example of a brand that uses its loyalty card and mobile app to collect and identify customers. Records purchases, including where and when they were purchased. Starbucks uses predictive analytics to provide personalized messages to customers.
   It includes suggestions and special offers to increase the customer’s average order value when the user visits the local store.
4. Nike — Last year Nike launched a new system that allowed customers to design their own sneakers in store. Not only is this a great gimmick to drive sales, but it also collects a huge amount of useful data that machine learning algorithms can use to design future products and deliver personalized recommendations and marketing messages.
5. eBay Uses Brand Language Optimization to Drive Email Marketing Success — The company has been working with the artificial intelligence-powered customer experience platform Phrasee to enhance its marketing copy, focusing on email. Which uses a combination of natural language generation and deep learning to create copy at scale while dynamically optimizing performance.

How AI is changing marketing and advertising..?

Artificial Intelligence is already changing marketing and advertising, and its widespread use is changing the way businesses interact with consumers. One of the most visible areas of AI impact is personalization, which involves tailoring advertising and advertising campaigns to consumers’ interests, preferences, and behaviors.

Artificial intelligence technologies such as machine learning algorithms and natural language processing (NLP) can help businesses analyze large volumes of customer data and generate better insights for targeting and messaging.
Another area where artificial intelligence comes into play is the use of chatbots, which are computers that can attempt to chat with a human user.

Predictive analytics is also gaining attention in marketing and advertising. This includes using machine learning algorithms to analyze customer data and predict future behavior such as which products they will buy or which channels they will like the most.

Potential drawbacks of AI in marketing and advertising

One of the biggest concerns is privacy. As AI gathers more and more information about customers and their behavior, this information can be misused or fall into the wrong hands. Data or algorithms used in AI can also be biased, leading to discrimination.
To mitigate these risks, companies must take steps to ensure their AI systems are transparent, ethical and secure.
This includes being transparent about what data is collected and how it is used, and giving consumers control over their data. Companies should regularly review their AI systems for bias and take steps to resolve any issues that arise.
Another drawback of having AI in marketing and advertising sector is the risk of over-dependance on technology. While AI can help companies multitask and make better decisions, it should not replace human thinking and creativity. Companies need to strike a balance between artificial intelligence and human intelligence techniques to achieve the best results.

The Future of AI in marketing and advertising

As artificial intelligence (AI) continues to evolve and evolve, its role in marketing and advertising will only expand. In the future, AI will play a larger role in these areas, providing new and innovative ways for companies to connect with customers and build brands.
One difference that will be important is voice search. As more users use Amazon’s voice-activated assistants like Alexa and Google Home, companies must optimize their content and ads for voice search queries. This will require a different approach to SEO and keywords, with a focus on creating content that is easy to understand and answer with voice.
Another area where AI can have a significant impact is augmented reality (AR). AR technology is already being used in advertising and marketing, allowing consumers to try on clothes or see how the furniture will look in their home. As AI becomes more effective, AR experiences will likely become more personalized and interactive, providing new ways for companies to engage with customers.

The AI is also expected to continue to improve targeting and identity. With its ability to analyze large amounts of data and make predictions based on that data, AI can help companies better understand their customers and deliver personalized content and experiences the way they want and like it.
However, there are risks and challenges which come along new oppurtunities . Companies need to be vigilant to make sure their use of AI is fair and transparent, and to protect consumer privacy. Additionally, AI can impact businesses and operations as it continues to become more automated and decision-making processes.
Overall, the future of AI in marketing and advertising is exciting and full of potential. By being aware of events and technologies, companies can use artificial intelligence to create effective and engaging campaigns while being aware of the risks and opportunities posed by this new technology.

Conclusion

As a result, increased efforts in marketing and advertising have a significant impact on the business. While this technology has some downsides and risks, the benefits are clear, including improved performance, improved customer experience, and better campaigns. As AI continues to evolve and evolve, it will become an even more important tool for marketers and advertisers, helping them prioritize and reach their target customers more effectively. Companies that embrace AI and use it responsibly and ethically are reaping the benefits, while companies that ignore the technology may find themselves lagging behind their peers. It is clear that artificial intelligence is permanent and its impact on business and advertising could be significant in the years to come.

References

[1] Ameen, N., Tarhini, A., Reppel, A., & Anand, A. (2021). Customer experiences in the age of artificial intelligence. Computers in Human Behavior, 114, 106548.

[2] Brynjolfsson, E., & Mcafee, A. N. D. R. E. W. (2017). Artificial intelligence, for real. Harvard business review, 1, 1–31.

[3] Brynjolfsson, E., Rock, D., & Syverson, C. (2018). Artificial intelligence and the modern productivity paradox: A clash of expectations and statistics. In The economics of artificial intelligence: An agenda (pp. 23–57). University of Chicago Press.

[4] Dwivedi, Y. K., & Wang, Y. (2022). Guest editorial: Artificial intelligence for B2B marketing: Challenges and opportunities. Industrial Marketing Management, 105, 109–113.

[5] https://www.analyticsvidhya.com/blog/2023/03/ai-marketing-strategies/

Authors

1. Omkar Karpe
2. Sakshi Ozarde
3. Ganesh Karode
4. Rutika Masane

Agile scrum artifacts are information that a scrum team and stakeholders use to detail the product being developed, actions to produce it, and the actions performed during the project. The main agile scrum artifacts are product backlog, sprint backlog, and increments.

The Three Scrum Artifacts

Scrum uses three artifacts to help manage work. And these are following:

Product Backlog — The product backlog is an ordered list of everything that is known to be needed in a product based on the product goal. It is constantly evolving and is never complete.

Sprint Backlog — The sprint backlog is a list of everything that the team commits to achieve in a given sprint. Once created, no one can add to the sprint backlog except the development team.

Potentially Releasable Product Increment — At the end of every sprint, the team delivers a product increment that is potentially releasable, meaning that it meets their agreed-upon definition of done. (An example might be fully tested and fully approved.)

Product Vision

A Scrum project mostly starts with a vision of the product or system to be developed. The product vision in Scrum is a brief statement of the desired future state that would be achieved by developing and deploying a product. A properly envisioned project provides a definitive path that can best fulfill the project’s objectives or goals. It also provides a common understanding of the direction they want to move towards. Besides that, the Product vision also supports the Product Owner in prioritizing what features to build first, in what order, and what not to build for the Product.

Sprint Goal

Sprint goal is a high-level summary of the goal the product owner would like to accomplish during a sprint, frequently elaborated through a specific set of product backlog items. A sprint goal is a short, one- or two-sentence, description of what the team plans to achieve during the sprint. It is written collaboratively by the team and the product owner.

Backlog

A well-prioritized agile backlog not only makes release and iteration planning easier, it broadcasts all the things your team intends to spend time on — including internal work that the customer will never notice. This helps set expectations with stakeholders and other teams, especially when they bring additional work to you, and makes engineering time a fixed asset.

Product Backlog

A product backlog is a prioritized list of work for the development team that is derived from the roadmap and its requirements. The most important items are shown at the top of the product backlog so the team knows what to deliver first. The development team doesn’t work through the backlog at the product owner’s pace and the product owner isn’t pushing work to the development team. Instead, the development team pulls work from the product backlog as there is capacity for it, either continually (kanban) or by iteration (scrum).

Sprint Backlog

The Sprint Backlog is a plan by and for the Developers. It is a highly visible, real-time picture of the work that the Developers plan to accomplish during the Sprint in order to achieve the Sprint Goal. Consequently, the Sprint Backlog is updated throughout the Sprint as more is learned. It should have enough detail that they can inspect their progress in the Daily Scrum.

Components of Sprint Backlog

• User story: features written from the perspective of the customer.
• Task name: name of each sprint task.
• Task description: brief description of each sprint task.
• Task prioritization: priority of each task relative to each other
• Sprint burndown chart: graph of work left to do vs time to complete it.
• Daily time allocation: amount of time it takes to complete each task.

Burndown Chart

What is a Burndown Chart?

During sprint meetings, teams determine the work breakdown of the project and predict the time in which each task can be completed. From this task breakdown, the plots of the burndown chart can be created.Once created, a line is shown to reflect the ideal number of effort hours needed to complete the project. As the project progresses, teams can use the burndown chart to:

• Determine the amount of work done in each iteration
• Show the work completed
• Visualize the remaining work
• Predict when a project will be finished

A burndown chart in software development shows the progress of ‘burning down’ the pile of remaining work.

In short, a burndown chart shows how much work remains to be done (y-axis) at any given day since work started (x-axis) and until the work is complete.

Author:

Rohit Jadhav

Omkar Karpe

Ganesh Karode

Sushil Kandhare

MS SQL Server History

MS SQL Server History started with the first product; SQL Server 1.0. It was a 16-bit widescreen server for OS / 2 in 1989. The history of Microsoft SQL Server continues even today.

In 1987, Microsoft partnered with Sybase Solutions to develop a website management system that could compete with existing giants such as IBM and Oracle.

It was determined that Sybase would have all the commercial rights and benefits derived from the Product version of the Non-Microsoft forum while Microsoft will have exclusive copyright to the Web site developed by Microsoft Platforms.

So the first product of the website server was released in 1989. Later, Sybase sold all the rights to Microsoft and now the brand name has been changed to MS SQL Server. To date, about 15 versions of this product have been released.

From these 15 here are the versions that are still supported and can be utilized by you:

1. MS SQL Server 2012

2. MS SQL Server 2016

3. Azure SQL Database(latest version)

4. MS SQL Server 2017 (latest version)

5. And, MS SQL Server RC 2019 (preview version)

Introduction:

In today’s world, for data to be useful, it needs to be organized, efficient, timely and accurate. Almost all information sites provide reporting functionality as a closely related aspect. This project focuses on producing web-based reports using MS SQL Server and MySQL. This case study will help to acquire and improve information about Web-based reporting on websites.

Microsoft provides a server-based reporting platform to generate comprehensive reports using a variety of data sources.

SQL Server is an affiliate database management program, or RDBMS, developed and marketed by Microsoft.

SQL Server has been operating exclusively for Windows environments for over 20 years. In 2016, Microsoft made it available on Linux. SQL Server 2017 was usually available in October 2016 running on both Windows and Linux.

Microsoft SQL Server is a market-leading business solution that is used by a large number and variety of applications where their data is stored and stored. Microsoft SQL Server is an incredibly powerful, scalable and robust solution, yet its durability often leads customers to the wrong idea of ​​security.

Uses of the MS SQL:

● To create databases.

● To maintain databases.

● To analyze the data through SQL Server Analysis Services (SSAS).

● To generate reports through SQL Server Reporting Services (SSRS).

● To carry out ETL operations through SQL Server Integration Services (SSIS).

SQL Server Components

The SQL server operates in a client server configuration, which is why it supports two types of components − (a) Workstation and (b) Server.

● Workstation components are installed on all SQL Server devices / machines. These are just interactive areas for Server components. Example: SSMS, SSCM, Profiler, BIDS, SQLEM etc.

● Server components are installed on the central server. These are services. Example: SQL Server, SQL Server Agent, SSIS, SSAS, SSRS, SQL browser, SQL Server for full text search etc.

Sql server Architecture :

Diagram:

SQL Server consists of two main components:

1. Database Engine
2. SQLOS

Database Engine

The core component of SQL Server is Database Engine. Database Engine contains a related search engine and a search engine that retains web files, pages, pages, indexes, etc. Website features such as saved processes, views, and triggers are also created and used by the Web Engine.

1)Relational Engine

Relational Engine contains components that determine the best way to use the query. Communication engine is also known as query processor.

The related engine requests data from the backup engine based on the input query and analyzes the results.

Other related engine functions include query processing, memory management, thread and task management, bath management, and distributed queries.

2)Storage Engine

The storage engine handles the storage and retrieval of data from storage systems such as disks and SANs.

SQLOS

Below the engine associated with the last engine is the SQL Server Operating System or SQLOS. SQLOS provides many operating system resources such as memory and I / O management. Other services include separate administration and synchronization services.

Advantages of Instances

● To install different versions in one machine.

● To reduce cost.

● To maintain production, development, and test environments separately.

● To reduce temporary database problems.

Pros of Microsoft SQL Server:

1. Increases data security

One of the main goals of Microsoft SQL Server is to ensure the security of your site, especially through the Microsoft SQL Server website management service. This software allows you to work with a table structure that integrates tasks and data elements, helping to protect the data you own. With customer information and other sensitive information, site security and integrity are required

2. Optimized data storage

With Microsoft SQL Server, you do not have to have another data storage on the same website if you are using a different device. This allows you to manage data easily and efficiently with problem-solving and minimal maintenance. As a result, you can save time and work on other important aspects of your business.

3. Data recovery support

In the event of a power outage or server shutdown, the data may be corrupted, which poses a serious problem for businesses with little or no backup. The Microsoft SQL server eliminates the risk of data loss by having data recovery features. As a result, you will have more peace of mind knowing that your data is protected by caching, log files, and common backups, no matter what may happen to your server.

The cons of Microsoft SQL Server:

1. Restricted compatibility

Making additional investments in Microsoft software may be needed if your business uses little to no Microsoft infrastructure. These extra commitments might also cost your company more, but they will allow you to use Microsoft SQL Server on your platform.

2.Cost

There is a free version of Microsoft SQL Server that you can use. But with highly advanced data and features applications, you will need to make huge investments in high-end software versions.

Reference:

https://www.sqlservertutorial.net/getting-started/what-is-sql-server/

https://www.tutorialspoint.com/ms_sql_server/index.htm

https://www.geeksforgeeks.org/advantages-and-disadvantages-of-sql/

https://en.wikipedia.org/wiki/Microsoft_SQL_Server

By-

CS-B, Batch-3, Group-SY35

Authors-

Omkar Karpe

Ganesh Karode

Sushil Khandare

Rohit Jadhav

The Post-Turing Machine is a very simple type of Emil Post’s Turing Machine-equivalent calculation model.

The Shipping Machine works by first reading the program, which ends as a string list, and then obtains the user input in the form of a series.

HISTORY

In 1936 he introduced the Turing Machine, and shortly thereafter Emil Leon Post (1897–1954) invented the postal machine.

They hoped it would become a “Universal Algorithm”. The condition that must be met with “such an algorithm” is that any language that can be accurately described by People, must be adopted by another version of this machine.

This can make it much more powerful than the FA or PDA

1.

Finite Combinatory Processes — Formulation 1

Post’s model of a computation differs from the Turing-machine model in a further “atomization” of the acts a human “computer” would perform during a computation.

1936

2.

Recursive Unsolvability of a Problem of Thue.

Atomized the Turing 5-tuples to 4-tuples:

“Our quadruplets are quintuplets in the Turing development. That is, where our standard instruction orders either a printing (overprinting) or motion, left or right, Turing’s standard instruction always order a printing and a motion, right, left, or none”

1947

3.

Wang model.

Reduction to only four instructions of the Wang model (Wang B-machine) as the source of the “program formulation” of binary-tape Turing machines using numbered instructions from the set {write 0, write 1, move left, move right, if scanning 0 then goto instruction i, if scanning 1 then goto instruction j}.

1957

4.

Davis model (Turing–Post machine).

Davis assigns the numbers “1” to Post’s “mark/slash” and “0” to the blank square. To quote Davis: “We are now ready to introduce the Turing–Post Programming Language. In this language there are seven kinds of instructions: {PRINT 1, PRINT 0, GO RIGHT, GO LEFT, GO TO STEP i IF 1 IS SCANNED, GO TO STEP i IF 0 IS SCANNED, STOP}.

1978

5.

Davis–Sigal–Weyuker’s Post–Turing program model.

This model allows for the printing of multiple symbols. The model allows for B (blank) instead of S0. The tape is infinite in both directions. Either the head or the tape moves, but their definitions of RIGHT and LEFT always specify the same outcome in either case (Turing used the same convention).

1994

WHAT IS TURING MACHINE ?

The Turing Machine was invented by Alan Turing in 1936.

The Turing machine contains a tape of unlimited length at which literary work can be done. Tape contains endless cells where each cell contains an insertion mark or a special marker called blank. It also contains a header pointer that identifies the current read cell and can navigate in both directions.

EXPRESSING TURING MACHINE

• A Turing Machine is presented as a 7-tuple (Q, T, B, ∑, δ, q0, F) where:
• Q is a set of limitations
• T is the alphabet of tape (symbols that can be written on Tape)
• B is an empty symbol (all cells are filled with B except input characters)
• ∑ Input alphabet (symbols that are part of input characters)
• Δ is a transition function that sets the map Q × T → Q × T × {L, R}.
• q0 is the first condition
• F a set of endpoints. If any F status is reached, the input unit is accepted.

WORKING

PM’s accept some context-free languages and some non-context- free languages.

For any Turing Machine there exists a Post Machine which accepts the same language, and vice versa. The proof that for any Post Machine there exists a Turing Machine, say M, which accepts the same language, follows from the fact that the queue of the Post Machine can be represented by the leftmost part of the tape of the Turing Machine M and the operations on the queue can be simulated by operations on the tape.

A Turing machine consists of an infinitely long tape, which has been divided up into cells. Each cell can contain either a 1, a 0, or an empty space. Above one cell of the tape is a head, which can either move left or right, and can read the symbols written in the cells. The head is also capable of erasing symbols and writing new symbols into the cells.

OPERATIONS

There are three different tasks the Shipping Machine can complete:

O1: The machine goes into an unusable state: you must label the cell that already has the label, or delete the cell without the label. Then the execution is stopped And an ineffective suspension occurs.

O2: The machine comes to a command setting. The machine has completed the process.

O3: The machine never stops or comes to an unusable command: it goes into an endless loop.

PROPERTIES

Alphabet for input letters and special mark # (aabb #). A linear end is called a store or line containing a complete input unit (using the FIFO stack). This area can be read, which means that the left-hand character can be removed for testing.

STORE can be added as well, which means the new character can be connected to the right of anything that already exists.

Read the provinces — which removes the left-hand character from the SHOP and branch accordingly. The end of the branch on the machine occurs in READ provinces. Branch which means empty Store has been read. PMs determine, so no two edges from READ have the same label.

Add states — which concatenate a character onto the right end of the string in STORE. This is different from PDA PUSH states, which concatenate characters onto the left. Post machines have no PUSH states. No branching can take place at an ADD state.

A START and some halt states called ACCEPT or REJECT — If we are in a READ state and there is no labeled edge for the character we have read then we crash, which is equivalent to taking a labeled edge into a REJECT state. We can draw our PM’s with or without REJECT states.

Example,

Post machine is deterministic. So, we do not have the edges that lead to REJECT states but instead we allow the path to crash in the READ state if there is no place for it to go. {a^n, b^n}

REFERENCES

1. https://www.researchgate.net/publication/264301619_Turing_Machine_and_Automata_Simulators
2. https://upscfever.com/upsc-fever/en/gatecse/en-gatecse-chp144.html
3. https://www.youtube.com/watch?v=uvqh0EToQSQ
4. https://www.wolframscience.com/prizes/tm23/images/Post.pdf
5. https://www.wolframscience.com/prizes/tm23/images/Post2.pdf
6. https://en.wikipedia.org/wiki/Wang_B-machine
7. https://thtsearch.com/content/Post%E2%80%93Turing_machine/
8. https://tinman.cs.gsu.edu/~raj/8910/f18/DavisSigalWeyukar.pdf

By-

CS-B, Batch-3, Group-SY35

Authors-

Omkar Karpe

Ganesh Karode

Sushil Khandare

Rohit Jadhav