Leaderboard

Good morning, Some of you have certainly noticed a change on the logo. Since the 2 last versions, we have set out to put generative artificial intelligence in the application with GPT and OPENAI. The module will certainly evolve in time to integrate other approach like Google Bard for example. We go there in stages. For the next version 3.431 a Charbot will be integrated into the application on the front office. It will use generative AI to give additional information to the user allowing him to have more information on the product. We wanted to go more far including generation conditions for example. But the expected result was not there. We have tried to use text segmentation to reduce the processing time but the limitation obtained is not good. This is due to the fact that Mariadb does not have a feature to include a vector inside a table. This approach improves treatment time but drastically reducing the number of Token sent to GPT and the cost linked to treatment. In addition, proposing a solution whose reaction time and a cost of treatment too much did not seem to us to be careful as an approach. Furthermore, like the administration where we use Davinci-003 model, in the front office, we use GPT 3.5 Turbo is a model largely cheaper than Davinci. We have included a parameter allowing to manage the chatbot display in the offering front so that you can better manage your budget. To complete the next evolution will have updates concerning the GPT including new statistics, a preditive security stock -analyse and an update on customer recommendations. Our objective is to be able to offer a new generation of e-commerce platform allowing merchants (small company, medium company) to be able to use artificial intelligence with a budget control, to be able to carry out more impactful communication campaign, marketing and finally to optimize production time, creation of product sheets and increase the sell. The work continues and will continue in this orientation in the future.

In the fiercely competitive world of e-commerce, delivering a personalized and delightful shopping experience to customers is no longer just an option; it's a necessity. As an e-commerce business owner, you understand the importance of staying ahead of the game and providing your customers with product recommendations that align perfectly with their preferences and needs. Enter ClicShopping AI cutting-edge Product Recommendations System, a game-changer that enhances customer engagement, boosts conversions, and increases your sales. Understanding the Power of Product Recommendations: Product recommendations are like a knowledgeable shopping assistant, guiding your customers through a curated selection of products that are relevant to their interests. By leveraging data insights, customer behavior, and sentiment analysis, ClicShopping AI product recommendations System transforms ordinary shopping into an exceptional, personalized journey. The Magic of Sentiment Analysis: At the heart of ClicShopping AI product recommendations System lies sentiment analysis, a revolutionary technology that understands and interprets the sentiment or emotion behind customers' interactions. By analyzing product reviews, feedback, and customer interactions, sentiment analysis reveals the true sentiment - positive, neutral, or negative - customers associate with your products. Your Customers' Opinion Matters: With sentiment analysis, you gain valuable insights into your customers' preferences and opinions about your products. This enables you to identify products that resonate with customers and those that might need improvement. Armed with this knowledge, you can make data-driven decisions to enhance your product offerings and exceed customer expectations. Personalized Recommendations for Every Shopper: Say goodbye to generic product recommendations! ClicShopping AI advanced algorithms combine sentiment analysis with other factors like average ratings, user feedback, and sales data to create personalized recommendations for each individual shopper. By tailoring product suggestions based on their preferences, you foster a sense of exclusivity and make customers feel valued. Boost Conversions with Smart Recommendations: Imagine a scenario where a customer shows interest in a product but hasn't yet made a purchase. ClicShopping AI product recommendations System steps in with intelligent cross-selling and upselling recommendations that complement their chosen item. The result? Increased average order values and higher conversion rates. Seize the Opportunity with Best-Selling Products: Identifying your best-selling products becomes effortless with ClicShopping AI product Recommendations System. By leveraging data from sales analytics, you can confidently showcase these high-performing products to capture the attention of potential customers and drive more sales. Empower Customers with User Feedback: Customers trust the opinions of their peers. By incorporating user feedback into your product recommendations, you empower shoppers to make informed decisions. Positive feedback elevates a product's desirability, while addressing negative feedback shows your commitment to customer satisfaction. Conclusion: In a rapidly evolving e-commerce landscape, providing customers with personalized, engaging shopping experiences is paramount. ClicShopping AI product Recommendations System takes the guesswork out of product suggestions, enabling you to connect with your customers on a deeper level and boost your sales like never before. Embrace the power of sentiment analysis, data-driven insights, and intelligent recommendations, and watch your e-commerce business thrive in an era of customer-centricity and personalization. with ClicShopping AI, your customers' satisfaction is just a recommendation away!

In the dynamic world of retail, managing your inventory efficiently is a key factor in ensuring business success. Introducing the predicitive stock class from ClicShopping, a powerful tool designed to help beginners navigate the complexities of inventory management and optimize stock levels with ease. Understanding the predicitve Stock The predictive stock class is a remarkable addition to the ClicShopping AI ecosystem, offering a set of functions tailored to assist retailers in calculating safety stock levels based on historical demand data, lead time, and desired service levels. It's a behind-the-scenes hero that empowers you to make informed decisions about your stock, ensuring you never run out of products when your customers need them the most. Accurate Safety Stock Calculations Made Simple One of the core features of the predicitive stock is its ability to calculate safety stock levels. Safety stock acts as a buffer, protecting your business from unexpected fluctuations in demand and lead times. With just a few lines of code, you can harness the power of statistical analysis to determine the optimal safety stock level for your products. The class leverages historical demand data to compute the mean (average) and standard deviation of product sales. These insights are then combined with the lead time and desired service level to generate a reliable safety stock figure. By adopting this approach, ClicShopping AI empowers beginners to make data-driven decisions, ensuring products are readily available for customers, even during peak periods. Personalized Configuration for Your Business ClicShopping AI understands that every business is unique, and stock management requirements can vary greatly. The predicitive stock offers customization options to suit your specific needs. You can adjust parameters such as lead time, service level, and standard deviation factor to fine-tune the safety stock calculations according to your business context. Seamless Integration and Ease of Use Integrating the predictive stock into your ClicShopping AI environment is a breeze. Whether you're a seasoned developer or just starting out, the class's well-structured functions and clear documentation make the implementation process straightforward. With a few lines of code, you can tap into the world of data-driven inventory management and take control of your stock levels like a pro. Real-World Application Let's explore a real-world scenario where the predicitive stock shines. Imagine you're an online retailer preparing for the holiday season—a time of heightened demand and potential supply chain disruptions. By utilizing the predicitive stock, you can analyze historical sales data, factor in lead times, and set a service level that aligns with customer expectations. The class then calculates the optimal safety stock, ensuring you have the right amount of inventory on hand to fulfill orders promptly, even in the face of unforeseen challenges. Conclusion ClicShopping's predicitive stock is a game-changer for beginners entering the realm of inventory management. It empowers you to make informed decisions, eliminate stockouts, and provide exceptional service to your customers. By harnessing the power of statistical analysis, you can confidently navigate the complexities of stock optimization and take your business to new heights. Are you ready to transform your inventory management? Embrace the predicitive stock from ClicShopping and unlock a world of data-driven success. Experience the future of retail with ClicShopping today.

Like this article and very pertinent

@Gary Just a little rectification about Gary post. Not composer install mylibraries but composer require mylibraries

Hello @Speedy, Yes there is a limitation about the response when the tokens number is reach. You can change inside the #GPT setting the number of token you want to use for every request. You can customize your setting and test it until your are satisfied. It's important to understand the concept, temperature, token . The response can change a lot.

@ClicShopping, Thank you for this update, yes work for me now @GuillaumeSochon look this error in your log [23-Aug-2023 03:18:24 America/New_York] PHP Fatal error: Uncaught PDOException: SQLSTATE[42S02]: Base table or view not found: 1146 Table 'sogu7276_clic451.csw2_marketplace_file_informations' doesn't exist in /home/sogu7276/shop/includes/ClicShopping/OM/Db.php:107 You must applied this patch in your phpMyadmin. Before the latest release marketplace_file_informations inside schema directory was not correct, a patch has been applied and resolve this problem. CREATE TABLE `clic_marketplace_file_informations` ( `id` int(11) NOT NULL, `file_id` int(11) NOT NULL, `file_name` varchar(255) DEFAULT NULL, `date_created` date DEFAULT NULL, `date_updated` date DEFAULT NULL, `file_version` varchar(255) DEFAULT NULL, `file_downloads` int(11) NOT NULL, `file_rating` int(11) NOT NULL, `file_prices` decimal(15,4) DEFAULT NULL, `file_date_added` date DEFAULT NULL, `file_url_screenshot` text DEFAULT NULL, `file_url_download` text DEFAULT NULL, `is_installed` tinyint(4) NOT NULL ) ENGINE=InnoDB DEFAULT CHARSET=latin1 COLLATE=latin1_swedish_ci; -- -- Index pour les tables déchargées -- -- -- Index pour la table `clic_marketplace_file_informations` -- ALTER TABLE `clic_marketplace_file_informations` ADD PRIMARY KEY (`id`), ADD KEY `index_file_id` (`file_id`); -- -- AUTO_INCREMENT pour les tables déchargées -- -- -- AUTO_INCREMENT pour la table `clic_marketplace_file_informations` -- ALTER TABLE `clic_marketplace_file_informations` MODIFY `id` int(11) NOT NULL AUTO_INCREMENT; ## patch ALTER TABLE `clic_marketplace_file_informations` CHANGE `is_installed` `is_installed` TINYINT(4) NOT NULL DEFAULT '0'; ALTER TABLE `clic_marketplace_categories` CHANGE `sort_order` `sort_order` INT(3) NULL DEFAULT '0'; ALTER TABLE `clic_marketplace_categories` CHANGE `parent_id` `parent_id` INT(11) NOT NULL DEFAULT '0';

Hello @GuillaumeSochon, Welcome, i just change the title and remove your website link, just a better practice. I hope it's fine for you. About the Drack report, I will look but I recommend to open an issue on Github. If there is a bug it will be fixed for a next release. I looked the log and I see you, something happen with the api. After testing, it's important to update the sql request. It will be implemented on the next release. Do not forget after to actualize your page to see the information ALTER TABLE `clic_marketplace_file_informations` CHANGE `is_installed` `is_installed` TINYINT(4) NOT NULL DEFAULT '0'; ALTER TABLE `clic_marketplace_categories` CHANGE `sort_order` `sort_order` INT(3) NULL DEFAULT '0'; ALTER TABLE `clic_marketplace_categories` CHANGE `parent_id` `parent_id` INT(11) NOT NULL DEFAULT '0';

Hello, I suppose you want to install this app : catalog apps catalog import export - What is the version of ClicShopping AI do you have ? - First, I tested the marketplace and do not have error as you, I suppose inside the setting you make a mistake : Do you insert your email and password ? - I noted a bug on my side, the api do send the information, I think it's better to report this issue on github - If nothing work for you, you can install manually the application, download it and push the files inside your repository. - to install it,look the section install for this App. In your case it will be ......./index.php?A&Catalog\ImportExport. Little note. Do not include your website. Just the most important. You will be more safe again an attack. Do you read this tuto ? https://www.clicshopping.org/forum/blogs/entry/5-secure-clicshopping/

Hi, I am from Softaculous. We have asked Admin Email and Password as admin credential while install. You can Login through Admin Email and Password. If you still face issue you can open support ticket with us here : https://softaculous.deskuss.com/open.php We will be happy to help you.

Thank-You Patrick that worked. I installed using SOFTACULOUS which makes admin the default username. After your reply, I remembered seeing an email when I went to the CLICSHOPPING demo. Pietro

OpenAiGPT can help companies in online trade in several ways. Here are some examples : GPT and analysis Content generation: GPT can generate quality content for e-commerce websites, such as product descriptions, technical sheets and blog articles. This can help companies increase their online visibility and improve their natural referencing. Selling assistance: GPT can also be used to create virtual assistants to help customers find products and answer their questions. This can help companies improve customer experience and increase conversions. Data analysis: GPT can be used to analyze e-commerce data, such as sales trends, purchase behavior and customer comments. This can help companies make informed decisions on products for sale and the advertising campaigns to be launched. Generation of automated Chatbot script: GPT could help generate automated script for chatbot for better customer support and better shopping experience. Generation of slogan or catchy title: Chat GPT can also be used to generate slogans or catchy titles for online advertising campaigns. This can help companies attract customer attention and increase conversions. It is important to note that Chat GPT is a powerful tool but it does not replace creativity and human expertise. It is best to use it as a tool to help you improve the performance of your site and your customer experience Could GPT do the after -sales service of an online commerce site? ‍ GPT could certainly be used to manage the after-sales service of an online commerce site. It could be used to create a chatbot or a virtual assistant to help customers solve their problems and answer their questions. - It could also be used to analyze after-sales service data, such as support tickets and customer comments, to identify trends and recurring problems. - It could use pre-trained models on customer needs to be able to meet the most common requests. - It can help using standard expressions and responses for requests for return, refund, order, delivery monitoring etc. - It could also be used to generate scripts for customer service employees to help them respond to requests in an efficient and professional manner. It is important to emphasize that even if GPT has high capacities for language understanding, it is important to train it with data specific to your company, so that the generated responses are as appropriate as possible. It is also important to ensure the quality of these responses and check them regularly to ensure that they are correct and in accordance with the company's policies. Will it be possible to send emails directly to customers, in particular for informed of a delivery state for an order? ‍ - It would be possible to use Cat GPT to send automatic emails to customers to inform them of the state of their delivery. - This could be accomplished using the GPT API to generate automated emails, which would then be sent to customers via an email sending platform. For this, it would be necessary to configure an automated email generation system which will integrate into your order management system, so that it can recover the delivery data corresponding to a given command and use it to generate a personalized email to the customer. - GPT could then be used to generate the content of these emails, using delivery and shipping information to customize the message. - It would also be possible to use GPT to generate email models for different stages of the delivery process, such as order confirmation, shipping, delivery or return. This would automate the process for sending delivery monitoring and saving time for employees. It is important to note that the sending of automated email must respect the rules of confidentiality and spam and must comply with the applicable laws and regulations, such as the Data Protection Act (GDPR). It is important to ensure that customers have given their consent to receive communications of this type and give them the possibility of unsubscribing. Product content creation : How can GPT Chat be useful in their creation? ‍ GPT-4 can be very useful for the creation of product sheets for an online commerce site. It could be used to automatically generate product descriptions, technical characteristics, advantages and disadvantages, current uses, product images and even catchy titles for products. Here are some concrete examples of the use of Chat GPT for the creation of product sheets: - Generation of product descriptions: GPT could be used to generate convincing and attractive product descriptions, using information on the characteristics and advantages of the product. He can use examples of sales sentences to encourage the user to buy. - Generation of technical characteristics: GPT could be used to generate the detailed technical characteristics of products, such as dimensions, weight, materials used, etc. - Generation of product images: GPT could also be used to generate product images based on technical characteristics and product descriptions. He could use existing images based on tags for image recognition. - Generation of catchy titles: GPT could also be used to generate catchy titles for products, which would attract the attention of customers and encourage them to learn more about the products. - Generation of SEO content: GPT could also be used to generate SEO (Search Engine Optimization) content to improve product sheets in terms of natural referencing. It is important to note that even if GPT can automatically generate quality content and it is important to check and reread the product sheets before publishing them to ensure that they are correct and respect the rules of the company in terms of quality and presentation. It is also important to have complete information on products to feed GPT-4, to avoid errors or inaccuracies. How can GPT analyze the sales data of a website and what will interest be? ‍ GPT can be used to analyze sales data from an electronic commercial website using automated data analysis techniques. It could be used to identify sales trends, customer purchase behavior and the most popular products, for example. Here are some concrete examples of the use of Chat GPT for sales data analysis: - Identify sales trends: GPT could be used to analyze sales data and identify trends, such as periods of high activity or most popular products. This could help companies plan their inventory and adapt their marketing strategy. - Analysis of purchase behavior: GPT could also be used to analyze customer purchase behavior, such as products purchased together, purchase frequencies and spent amounts. This could help companies better understand their customers and adapt their marketing strategy accordingly. - Customer segmentation: GPT could use sales data to identify customer segments that have similar characteristics and similar purchase behavior. This could help companies target their marketing campaigns and personalize offers for each group of customers. - Sales forecasts: GPT could use machine learning techniques to predict to come from the historic sales data. This could help companies plan the inventory and marketing strategies. - Customer commentary data analysis: GPT could use feelings analysis techniques to extract positive and negative information from customer comments, to identify the strengths and weaknesses of the products and the purchasing experience . This could help companies improve their products and improve the purchasing experience. Analysis of sales data can provide businesses with valuable information on their market, customers and products. This can help them make informed decisions on products for sale, the advertising campaigns to be launched and the marketing strategies to use, to increase sales and improve customer experience. How can GPT help the purchase service from an online sales site? ‍ ‍ GPT can help the purchase service of an online sales site using its natural language and content generation comprehension capacities. It can be used to automate certain tasks and improve the efficiency of the purchasing service. Here are some concrete examples of the use of Chat GPT for the purchase service of an online sales site: - Supplier search: GPT could be used to carry out online supplier searches, using specific keywords and search criteria to find potential suppliers. It could also be used to assess the quality and reliability of potential suppliers. - Generation of purchase requests: GPT could be used to automatically generate purchase requests for products, using supplier information and business needs to personalize requests. It could also be used to generate quotes and contracts for suppliers. - Monitoring of orders: GPT could be used to monitor orders and to inform suppliers and internal employees about possible delays or delivery problems. It could also be used to generate reports on commands and deliveries. - Price negotiation: GPT could use its language understanding capacities to help buyers negotiate prices with suppliers, using relevant arguments and responding to objections. - Data analysis: GPT could use data analysis techniques to identify trends and purchase opportunities, using information on orders, deliveries, stocks and prices to make purchase decisions informed. By using GPT to automate certain tasks and to improve the efficiency of the purchasing service, companies can save time and resources while improving the quality and relevance of purchasing decisions. It is important to note that the use of Chat GPT for these tasks does not replace the need for human expertise to make crucial decisions for business management, especially for negotiations and data analyzes. What can be the use of GPT chat for images produced from an eCommerce site? ‍ GPT can be very useful for product images of an electronic commerce site using its image recognition and content generation capacities. It can be used to automate certain tasks related to products and to improve the user experience. Here are some concrete examples of the use of GPT-3 for product images of an electronic commerce site: - Image recognition: GPT could be used to identify the characteristics of product images, such as colors, shapes and sizes, using automated image recognition techniques. This information could be used to improve product search or to suggest similar products. - Image generation: GPT could be used to generate product images based on product descriptions, using technical characteristics and product descriptions to generate images that faithfully represent products. - Image compression and resizing: GPT could be used to optimize the size of product images to improve the loading time of the website pages. We hope these elements can be useful for you. GPT is disruptive tool and it's important to implement it if you want to improve your productivity.

ccording to a research by Accenture, AI could potentially boost profitability rates by an average of 38% across 16 industries by 2035. AI will particularly benefit the retail sector because of its immediate impact on two fundamental areas that have a serious impact in retail ecommerce; creating a very personalised shopping experience and making smart business decisions. Artificial intelligence in ecommerce is also helping ecommerce platforms to cater the needs and interest of the users. 1. AI-Assisted Product Recommendations Retailers are able to use Big Data (the collection and compilation of data from different channels, previous transactions and other activities) to determine the trends in the shopping habits. This analysis could be used for personalized product recommendation. 2. Advanced Semantic-Based Search When we talk about personalization in the context of ecommerce and customer support, right now there is nothing better than intelligent customer engagement at every point of contact. One of the most important aspects of personalization is the ‘search result’. With advanced semantic-based site search, customers are able to easily navigate the store and find the relevant products without wading through a lot of unnecessary results 3. Chatbots Chatbots might just be the ultimate solution for the customer support woes. The good news is that modern chatbots are able to interact with customer databases and extract details of a customer and respond in the most appropriate and personalized way in real-time. Chatbot provides a comfortable customer experience that keeps visitors from leaving the website without making a purchase. 4. Virtual Assistants Other application of artificial intelligence in ecommerce is through virtual assistants. Virtual Assistants contribute a lot in creating an efficient sales process. These assistants communicate with customers to understand their preferences and provide them with an intuitive shopping experience. 5. Efficient Company-Wide Decision Making Pricing and Inventory Management Pricing products is no more a guess work. AI can use complex learning algorithms that assess market dynamics and all the other relevant factors of market competition. Optimal pricing of the merchandise helps a business stay relevant and ultimately succeed in a competitive environment. Business Analysis Putting all the data to use, you can also determine the kinds of products and business actions that have decreased efficiency, increased conversions, and if anything needs readjustments. These comparisons pertain to market share, pricing patterns and overall progress of a business. 6. Cybersecurity Enhancement Ecommerce websites have a ton of customer data that will be vulnerable if there aren’t any proper cybersecurity measures in place. Detection of New Threats Unlike traditional software, AI software tend to get smarter with time as new types of malware and threats emerge. Therefore, they will be able to quickly detect malware or ransomware attacks on your website before they’re able to cause any serious problems. Blocking Out Bots Bots form a large amount of the internet traffic today, and they can prove to be dangerous. Bots can steal credentials, create fake accounts and cause data fraud. 7. Sales Forecasting Sales forecasting and inventory management has become an absolute necessity for every business. To avoid any unforeseen shortages or stock run-outs, ecommerce businesses can now use artificial intelligence for sales forecasting. Software like Mintigo make use of AI to predict future sales based on consumer patterns and help decision-makers predict the results of their marketing efforts. 8. Fighting Fake Reviews Believe it or not, online reviews can make or break your brand. One fake review from your competitor or a bot can turn things around for your brand