A flight recommendation method based on multi-factor scoring

Abstract

The invention is related to a flight recommendation method based on multi-factor scoring, which provides recommendations to the user by considering many critical factors that affect the passenger's decision-making process, namely comfort score factor, flight price factor and safety score factor at the same time.

Description

[0001] A FLIGHT RECOMMENDATION METHOD BASED ON MULTI-FACTOR SCORING

[0002] Technical Field

[0003] The invention relates to a flight recommendation method based on multi-factor scoring that offers suggestions to the user by considering a large number of critical factors affecting the passenger's decision-making process at the same time.

[0004] State of the Art

[0005] The main technical problem present is the lack of solutions that give comprehensive and balanced flight advice, considering a large number of critical factors affecting the passenger's (or user's) decision-making process at the same time. Current flight booking and recommendation tools often prioritize uniform metrics such as price or duration, without sufficient consideration of vital considerations such as safety and comfort. This singular focus may lead to suggestions that do not match the various and subtle preferences of different passengers (or users). Today, passengers are looking for a more holistic approach to flight selection, where safety, comfort and affordable price are equally important. However, balancing these various factors in a single and consistent proposal solution brings about many difficulties. These challenges can be listed as follows:

[0006] 1. Various Data Sources and Formats: Safety, comfort and price data come from different sources and can be in various formats, which makes it difficult to integrate them into a combined solution.

[0007] 2. Subjectivity in Comfort and Safety Evaluation: Unlike the measurable price, comfort and safety are more subjective and more difficult to measure consistently.

[0008] 3. Dynamic Pricing and Quotas: Since flight prices change frequently, a system that can respond in real time is required to provide accurate recommendations.

[0009] 4. Personalization Needs: Passengers have unique preferences that require a system that can adapt to individual needs while maintaining an objective basis for comparison. In the present art, algorithms developed for comprehensive and user-friendly flight selection are of various shapes. When any prior art document or publication or patent document is examined, some of the previous methods and systems closest to the invention are as follows: a. Price-Oriented Comparison Tools: Many of the current solutions are primarily focused on comparing flight prices. These tools provide users with the opportunity to search for flights by cost, showing the cheapest options available. However, this method largely ignored other factors such as safety and comfort, which are vital for many passengers. This narrow focus can be cost-effective, but it can give recommendations that are not the best in terms of the overall travel experience. b. Single Metric (Measurement) Ranking Systems: Some solutions offer recommendations based on a single measurement, highlighting only one aspect of the travel experience, such as flight time or the number of stops (or stops). Although these tools offer more than a price comparison, they do not offer a holistic perspective on the flight experience. Therefore, these systems do not provide a comprehensive evaluation of what the flight experience requires and potentially miss the important factors that affect the passenger's decision. c. User Review Collection: Platforms that combine user reviews and ratings, which can also include comments on comfort and safety. However, these solutions are largely based on subjective feedback and do not offer a standardized method to compare flights according to these parameters. While collecting user reviews and ratings can provide valuable information, these solutions can be largely subjective and potentially biased, based on user-generated content. There is also a lack of standardization on how these ratings are applied to different flights, which makes it difficult to compare the options in the same way. d. Manual Research and Comparison: Before automated vehicles, some passengers manually research flights according to different criteria and need to compare them. This process is time-consuming and open to bias because individuals may not have access to comprehensive data or may give too much weight to certain factors. e. Collaborative Filtering: Although a popular approach in initial recommendation systems, collaborative filtering faces challenges such as the cold-start problem, where it becomes difficult to make accurate recommendations for new users or elements with little or no historical data. The meaning of this cold-start problem is that in systems using collaborative filtering, it is recommended based on the data of other users with similar information. But if a new user joins the system, the system will not have any data about him, so this is a problem because it cannot recommend anything to them based on other users. Because there is no data to be compared at the beginning. In addition, if there are a large number of users and elements here, the efficiency can be significantly reduced.

[0010] On the other hand, a flight recommendation method is mentioned in the CN111581506A - "Flight recommendation method and system based on collaborative filtering” document, which is another known state of the art. Although this method is a flight recommendation method, it is not a method that provides a recommendation to the user by including all three of the comfort, safety, and price data of the flight.

[0011] Although these solutions proposed in the previous art for flight advice solutions are innovative at that time, they bring many disadvantages as it is seen. Therefore, an algorithm (method) and system that eliminates the disadvantages of the present art, integrates into a composite score due to the variety of measurements, and allows personalization according to the priorities of the passenger should be developed.

[0012] Brief Description of the Invention

[0013] The object of the present invention is to realize a flight recommendation method (algorithm) based on multi-factor scoring that offers suggestions to the user by considering a large number of critical factors affecting the passenger's decision-making process at the same time.

[0014] Within the scope of this object, the invention is to improve and facilitate the flight selection process of the user (or passenger) by providing a more comprehensive, balanced, and personalized approach to flight comparison and advice. In summary, the invention also provides the following advantages: i. Balanced Evaluation of Flights: The invention provides a balanced evaluation of flight options by considering safety, comfort, and price issues, which are vital for the general travel experience but are generally overlooked in traditional flight comparison vehicles. ii. Personalized Flight Recommendations: It allows passengers to personalize their flight searches according to their personal preferences. By adjusting the weights of different measurements (safety, comfort, price), users can adapt the search results to better fit their priorities. Here, data is retrieved from open-source websites and airlines' websites. Price rates (or flight prices) do not need to be recorded in a database because they are taken from the internet and the airline's website in real time and change frequently. Comfort and safety do not change frequently and are not taken in real time but are instead updated periodically and stored in a database. iii. Simplified Decision-Making Process: The invention simplifies the flight selection process for travelers by providing a single, composite score that easily compares different flight options. The object of this approach is to make the decision-making process faster and more user-friendly. iv. Increasing Passenger Satisfaction: The algorithm of the invention increases general passenger satisfaction by offering more relevant and more suitable flight options to the passenger. This is achieved by considering a wider range of factors that affect the travel experience. In other words, most, if not all, systems recommend flights based on price. The method, which is the subject of the invention, accepts comfort and safety as new criteria for suggestions and offers more options to personalize the suggestions according to the needs of the users. v. Data-Based Insights for Industry Optimization: Providing airlines and travel agencies with valuable information on consumer preferences allows companies and agencies to optimize their offers and marketing strategies in line with current trends and customer demands. vi. Innovation in Travel Technology: The invention contributes to the advancement of travel technology by bringing a new approach to flight recommendations and sets a new standard that users can find what they expect from travel planning tools. vii. Adapting to Developing Passenger Needs: Meeting the developing needs of modern passengers who are looking for a more knowledgeable and holistic approach to planning their travels, considers not only the cost but also the quality and safety of their travel experiences.

[0015] Basically, the invention aims to revolutionize the way passengers choose flights, to make the process more compatible with their individual needs and preferences and thus to increase the overall efficiency and satisfaction of travel planning.

[0016] The flight recommendation method based on multi-factor scoring, which is the subject of the invention for flight recommendations, solves many technical problems in the field of travel planning and flight selection. The main problems it addresses and the solutions it brings: a. The solution to the lack of comprehensive flight evaluation tools (Integrated MultiFactor Evaluation): Unlike recommendation systems where traditional flight recommendation systems primarily focus on a single issue such as price or flight time, the method of the invention combines multiple basic factors (safety, comfort and price) into a single composite score (integrating into a single score), provides a holistic evaluation of flights and thus offers a more comprehensive evaluation of flights suitable for various priorities of passengers. The method combines three basic travel elements to provide a holistic evaluation of flight options. These elements are safety, comfort, and price. The method essentially uses a new method to analyze each flight based on these three different measurements. The Safety Score evaluates the safety information of the flight. The Comfort Score evaluates the flight experience. The Flight Price represents the cost of the flight. These metrics or measurements are evaluated separately and then integrated into a single and comprehensive score that reflects the overall value and attractiveness of each flight. This comprehensive evaluation is more in line with the different priorities of travelers and offers a versatile view for each flight option. In the invention, the safety scores and comfort rates (or scores) of each airline are taken from public sources. Safety rates (or scores) are above 7 points and represent the following:

[0017] IOSA audits the availability from world organizations such as ICAO country audit and EU and FAA bans. Past Death Reports

[0018] Past Accident Reports

[0019] Comfort rates (or scores) are calculated on the basis of 6 points according to the presence of the following:

[0020] Seat spacing (has more)

[0021] Blankets and pillows

[0022] In-flight entertainment

[0023] Website information

[0024] Beds

[0025] Meals

[0026] Safety scores are stored in a database for each airline. Each data source may have a different score range. When making recommendations, each rate is scaled (or normalized) to a value between 0 and 1. Users only choose weights (how important this measurement is to them). Then, in the method of the invention, a composite score is calculated by multiplying each of them by a predefined weight (selected by the user) and using the rate (comfort, safety, and price). According to the security score in the database and the weights from the user, these are transmitted to the algorithm and the algorithm creates a composite score with other metrics.

[0027] It is necessary to emphasize this important issue here. That is to say, the method of the invention can accept any rate related to comfort or safety, because the raw values from all sources are not used in the invention, they are scaled to values between 0 and 1. Thus, rates are converted to similar scales, making our approach more consistent and can adapt to different rates from different sources.

[0028] In this context, the invention provides a consistent and objective method to evaluate subjective aspects such as comfort and safety by standardizing subjective metrics. This systematic approach to scoring subjective factors is a significant improvement compared to relying solely on user reviews or inconsistent criteria. An innovative flight recommendation algorithm is presented here, which evaluates and ranks flights using a unique composite scoring method. The standardization, that is, the objectivity, is provided as follows:

[0029] Data on comfort score are collected, such as seat spacing and other comfort factors that do not require user feedback. Data on safety scores, such as inspections and accident reports, are also collected. This data does not require any user information or review. The comfort and safety scores of each airline are evaluated according to the presence of such services / reports in the airlines. Instead of using user feedback, it can objectively measure comfort and safety scores by focusing on metrics that can be measured (according to service presence) and thus avoid subjectivity. That is, passenger satisfaction is not measured here, and user comments are not used. Because these vary depending on people's opinions. Instead, the presence of services and comfort vehicles such as food, seat spacing, and pillows is measured. These are not subjective and are not based on the user's opinion, these services either exist or do not. b. The solution to the difficulty of balancing different passenger preferences: Despite the different preferences of travelers and not adapting to personalization in terms of prioritizing different flight issues in existing systems, the algorithm of the invention allows users to adjust the weights of each metric (safety, comfort, price), allowing personalized flight recommendations that meet the needs of individual passengers. Unlike traditional systems that focus on a single point, this algorithm provides a more comprehensive view of flight options by balancing multiple critical factors. Balancing here means that the user can choose the degree of giving importance to certain factors, which is personalized to them. These grades (i.e. weights) are then used in the composite score calculation where the importance of each factor is represented by these weights / grades.

[0030] Here, the user makes the weight / degree of each metric as follows:

[0031] The user can choose which factor is important for them, as well as the degree of importance of each factor for them. For example, a user may choose to focus only on the price factor. Or a user may want to focus on the three factors we have (price, comfort, flight fee). The user can choose which factor to consider: For example, the user should choose the flight price at 40%, safety at 40%, comfort at 20%. These are integrated and used in the method to determine the best flights for the user. In addition, if the user does not have any preferences, they can be pre-defined by the system. c. The solution to the inefficient and time-consuming flight selection: The algorithm of the invention simplifies the selection process, saves time, and reduces the decision-making fatigue of passengers by providing a single, easy to understand composite score for each flight, considering the comparison of flights on multiple platforms and various factors. d. The solution to the problem of adapting to dynamic market conditions (Dynamic and Real-Time Data Use): It is known that a recommendation system that can adapt to these changes in real time is required because flight prices and conditions are constantly changing. Therefore, it is necessary to offer the ability to combine real-time data to ensure that flight suggestions are always up-to-date and relevant. The algorithm of the invention is dynamic, the scores can be updated according to real-time data, and the relevant update is made to ensure that the suggestions remain accurate and relevant. Therefore, the dynamic nature of the algorithm allows it to adapt to changing market conditions and prices, which is a critical issue in the ever-changing travel industry. The algorithm uses real-time data for the price factor, provided that it accesses APIs and airline sites. The comfort and safety scores are updated internally in the database periodically, as they do not change very often. The term "dynamic" in the algorithm refers to the ability to adapt, update and respond to real-time and periodically changing data. e. The solution to the lack of customization (Customizable Weighting System) in flight search tools: Although existing flight search tools offer limited customization options, the customizable (or customizable) structure of the algorithm of the invention allows users to define their searches according to what is most important to them, resulting in more targeted and appropriate flight options. It allows users to personalize their flight searches by adjusting the weights of different metrics such as safety score, comfort score and flight price according to their personal preferences. This level of personalization is not commonly found in standard flight comparison tools and addresses the needs and priorities of individual passengers. This feature makes the method and system of the invention highly adaptable and user-friendly by meeting the various needs of different passengers.

[0032] In summary, the invention addresses the need for a more comprehensive, customizable, and user- friendly approach to flight advice. The invention simplifies the process of balancing multiple flight directions, personalizing search criteria and decision-making and thus increases the overall efficiency and satisfaction of the flight selection experience of passengers. On the other hand, the invention provides a user-friendly interface and allows the user to make a simple decision. Thus, the method (or algorithm) of the invention simplifies the flight selection process by providing an easy-to-understand score to the user through the interface and helps to make faster and more informed decisions. The algorithm can be integrated into a website / system that focuses on the user experience, which makes it accessible and understandable for a wide range of passengers. The invention provides data-driven insights for industry optimization, providing valuable information on consumer preferences and trends that will be useful for airlines and travel agencies.

[0033] The algorithm can serve as a tool for market analysis and strategy development in the travel industry. In addition, the invention offers scalability and versatility, allows for scalability, and can be adapted to other travel-related recommendations. Its versatile framework can be expanded beyond flights, which makes it applicable to various parts of the travel industry. With the continuous improvement potential of the invention, it can be continuously improved according to user feedback and changing travel trends. It provides long-term suitability and effectiveness by supporting iterative improvements. These advantages and unique elements collectively position this invention as a significant advance in the field of travel technology. The invention eliminates existing gaps in flight recommendation systems and improves the overall travel planning experience for users by providing a more sophisticated, user-oriented, and adaptive tool.

[0034] The most important quantity that distinguishes the invention from the current method is the ability to balance different factors in a combined score. This score serves as a simple and easy to understand measurement for passengers and helps them make informed decisions according to their personal preferences and priorities. The method, which is the subject of the invention, is designed to be user-centered and adaptive and allows users to see a clear sequence of flights according to this composite score. In addition, it offers users the option to adjust the emphasis of each measurement according to their individual travel needs and offers a highly customizable and versatile solution in flight selection.

[0035] The improvements of the multi-factor scoring approach in flight recommendations compared to other methods briefly include: i. Holistic Evaluation: It combines basic factors such as safety, comfort, and price in a single comprehensive score. ii. Balanced Evaluation: In contrast to systems that focus on one direction, it offers a more balanced perspective on flights. iii. Customizability: It allows users to adjust the weights of factors by adapting recommendations to personal preferences. iv. Objective Comparison: It reduces the dependence on subjective user data and provides a more objective basis for flight comparison.

[0036] Removes Limitations: The cold-initial problem encountered in collaborative filtering methods is not experienced in the method of the invention. In other words, user data that offers suggestions are not collected. This problem is present in systems that recommend flights based on the data of other users. In these systems, it tries to find users who are close to (have similar data with) a user to recommend flights to them, then suggests what other users close to them will choose. In other words, when a new user comes, since there is not enough information to find a user group close to them, a recommendation cannot be made to them. In contrast, in the method of the invention, user data is not used and factors that do not need user data are used, so there is no cold start problem in the invention.

[0037] The present invention offers many important advantages and uses similar to or much more than the above-mentioned advantages. These are: Scalability and Adaptability: The invention allows scalability and adaptability to other systems. This means that the method can be expanded to include additional measurements or can be adapted to other types of travel-related recommendations, such as hotels or package tours.

[0038] Improved Decision Making for Passengers: The system helps passengers make more informed decisions by providing a balanced and easy to understand composite score for each flight. This leads to increased satisfaction with travel choices and general travel experiences.

[0039] Increased Efficiency in the Flight Selection Process: The algorithm simplifies the flight selection process by providing a single, combined score for comparison, allowing users to save time and effort in researching and evaluating multiple flight options on different platforms.

[0040] Integration into Travel Agencies and Online Booking Platforms: The system of the invention has a significant potential for integration into existing travel booking platforms, increasing the value proposition to customers by providing a more comprehensive and personalized service that is not in the present art.

[0041] Data-Based Insights for the Airline and Travel Industry: The data collected and analyzed by this system provides valuable information on consumer preferences and trends. This will be useful for airlines and travel agencies to optimize their services and marketing strategies.

[0042] Continuous Improvement Potential: The system can be continuously improved or updated according to user feedback and changing travel trends. Thus, the long-term validity and effectiveness of the system subject to the invention is ensured.

[0043] In summary, the invention overcomes the challenges listed in the present art by not only integrating various measurements into a single, understandable score, but also creating an algorithm that allows personalization according to the priorities of the passenger. In response to many problems encountered in the present art, the multi-factor scoring algorithm of the invention offers a balanced and comprehensive approach to flight recommendation by integrating and normalizing various measurements such as safety, comfort, and price. In addition, it allows for personalization according to the user's preferences and makes it a more versatile and user- oriented tool. This new approach offers a more effective solution for passengers seeking versatile flight recommendations by overcoming the limitations of previous approaches. In this context, the invention offers a unique, user-oriented approach to flight recommendations and represents a significant advance in the field of travel planning and booking. As a result, a more special, reliable, and comprehensive flight recommendation vehicle emerges with the invention.

[0044] Descriptions of the Figures

[0045] Figure 1: It is the flow diagram of the steps of the method of the invention.

[0046] Description of References in Figures

[0047] The corresponding numbers in the figures are given below in order to better understand the invention:

[0048] 100. Method

[0049] 101. Entering the number of flights and flight details

[0050] 102. Entering at least three threshold values via the interface

[0051] 103. That the threshold can be pre-defined on the algorithm or selected by the user to personalize

[0052] 104. Calculating the final threshold

[0053] 105. Calculating the safety score by taking advantage of the accident rate, the inspections received and the death rates

[0054] 106. Calculating the flight price and comfort score metrics from the final threshold value

[0055] 107. Calculating the composite score

[0056] 108. Sorting according to the calculated composite score

[0057] 109. Displaying the result to the user through an interface

[0058] Detailed Description of the Invention The invention is related to a flight recommendation algorithm based on multi-factor scoring, which provides appropriate recommendations to the user in a short time by considering many critical factors that affect the passenger's decision-making process at the same time.

[0059] The technical interest or use of the multi-factor scoring algorithm for the flight recommendations of the invention lies in its innovative approach to solve the problem of complex flight selection for passengers. In general, the invention addresses an important gap in the travel industry by providing a more sophisticated, user-oriented, and effective tool for flight recommendations, reflecting the developing needs and preferences of modern passengers.

[0060] The system includes at least one user interface that offers suggestions by showing the data it receives / processes to the user. The data is transmitted to the user through a cloud technology through said interface.

[0061] The system of the invention comprises a device with at least one interface for at least one user to view the data stored in at least one database included in said system, at least one main server containing at least one database where said at least one device pulls said data and at least one communication unit that provides mutual wireless communication between at least one user device and at least one main server in order to operate the method of the invention and to make recommendations to at least one user.

[0062] Therefore, in order to realize the most appropriate recommendation in a short time by using the system of the invention, a flight recommendation method (100) (or algorithm) based on at least one factor of the invention includes the following working steps:

[0063] The user enters the number of flights and flight details through an interface (101)

[0064] The user uses three predefined thresholds or enters at least three thresholds for each factor (safety, comfort, flight price) via the interface (102) Said threshold values are set with at least one predefined (i.e. automatic) data collection / processing / integration unit or the user enters the threshold values manually (103)

[0065] Calculation of at least one final threshold value by the relevant data collection / processing / integration unit using the automatic threshold setting or manually entered threshold value (104)

[0066] At least one safety score metric, at least one accident rate, at least one audit and at least one death rate information are calculated by said data collection / processing / integration unit using THE threshold value selected or predefined by the user for the safety score (105)

[0067] Calculation of at least one flight comfort score and at least one flight price metric using the predefined THE threshold value for each factor by the relevant data collection / processing / integration unit (106)

[0068] Calculation of a composite score by the relevant data collection / processing / integration unit by combining the calculated safety rate value, comfort rate value and flight metric value (107)

[0069] Sorting of flights according to the composite score found (108)

[0070] Displaying the output of the sorted lists to the user through an interface (109)

[0071] To better explain the threshold expression given in the method steps, each factor has to have a threshold. If a threshold is selected by the user, the user must give a threshold for all factors. If the user wants to use the predefined threshold, the system will have the predefined threshold for all factors. Thresholds / weights are a way of measuring the severity of each factor, in other words, how much each factor will affect the results of the proposal. Here, each factor will have a threshold that represents its importance. Each factor has a weight / threshold value and will affect the results of the proposal, so the user can choose which one they care more and how much they care (1-100).

[0072] The data collection / processing / integration unit mentioned in the method of the invention also performs calculation and control operations according to the applications of the invention. These method steps are specific according to the following sub-steps:

[0073] At least one safety score mentioned in steps 105 and 106, at least one comfort score and at least one flight price, at least three data are collected in at least one database included in the system and these are integrated into the system.

[0074] The metrics such as the safety score, comfort score and flight price mentioned in steps 105 and 106 are normalized with at least one data collection / processing / integration unit on a common scale using the minimum-maximum normalization method.

[0075] The safety score, comfort score and flight price of the users mentioned in steps 105 and 106 are adjusted according to the weight preferences assigned to each metric and customizable weighting is made.

[0076] A composite score calculation is made in step 107 by multiplying each normalized metric by its corresponding weight and adding these values.

[0077] The flights are filtered and recommended to the user according to certain user criteria such as the ranking of the composite scores of the flights according to the flights that best fit the weighted preferences of the user, as well as the preferred airlines, flight hours or transfer preferences.

[0078] At least one dynamic and real-time data processing unit is operated to ensure that the latest prices and available information are reflected to the user through an interface and that the suggestions are always up-to-date.

[0079] The safety score includes a combination of at least one death score per airline, at least one accident score and at least one internationally recognized audit. Similarly, the comfort score includes at least one opportunity offered and at least one other in-flight comfort parameter. The flight price includes the price obtained from at least one airline database and at least one flight comparison website. The detailed description of the multi-factor scoring method for flight recommendations covers various components and elements that work together to create a comprehensive, user-friendly, and adaptive flight recommendation system. A breakdown of these elements and how they work in the invention is given below. In these steps, the relevant calculations and controls are performed by a data collection / processing / integration unit.

[0080] Step 1 : At least one safety score, at least one comfort score and at least one flight price, at least three data are collected with a data collection / processing / integration unit in at least one database included in the system and these are integrated into the system

[0081] Step 2: Since the metrics such as the safety score, comfort score and flight price in question are at different scales, the normalization of said metrics with at least one data collection / processing / integration unit on a common scale using the minimum-maximum normalization method

[0082] Step 3: Making customizable weighting by adjusting the weights assigned to each metric (Safety, Comfort, Price) of the users according to their preferences

[0083] Step 4: A composite score calculation is made by multiplying the relevant weight of each normalized metric and adding these values

[0084] Step 5: Ranking the composite scores of the flights according to the flights that best fit the weighted preferences of the user, as well as filtering and recommending the flights according to certain user criteria such as the preferred airlines, flight hours or transfer preferences

[0085] Step 6: Running at least one dynamic and real-time data processing unit to ensure that the latest prices and available information are reflected to the user through an interface and that the suggestions are always up-to-date

[0086] To give exactly what the common scale mentioned in Step 2 means; There is a different range of each factor in the resources available. For example, the range of the comfort score is 6, the range of the safety score is 7, and the price does not have a definite range. The way to normalize them is to gather the values of all factors under one common value. In the invention, each factor is normalized to a value between 0 and 1. These are normalized using the min- maximum equation. (y=(x-min) / (max-min)) . The details of these steps are given below:

[0087] Step 1- Data collection for three basic metrics: The invention includes 3 basic metrics: at least one safety score, at least one comfort score and at least one flight price. The safety score includes a combination of at least one death score per airline, at least one accident score and at least one internationally recognized audit. The Comfort Score includes at least one opportunity offered and at least one other in-flight comfort parameter. The flight price includes at least one flight price obtained from at least one airline database and at least one flight comparison website. The data sources such as at least one safety score, comfort score and flight price are integrated into a single system subject to the invention and it is ensured that each flight option is evaluated comprehensively.

[0088] Step 2: Since metrics such as at least one safety score, comfort score and flight price are at different scales, the metric is normalized to a common scale (e.g. from 0 to 1) using methods such as minimum-maximum normalization. This eliminates scale bias by ensuring that each metric contributes equally to the final composite score. For example, if the price factor (represents the real price) is 5000 TRY and the safety is 7 / 7 and the comfort score is 4 / 6, since the numbers are in a completely different scale / range, it is ensured that each score is normalized to a value between 0 and 1 and each factor contributes equally to the calculator since they will all be between 0 and 1.

[0089] Step 3: Users can adjust the weights assigned to each metric (Safety, Comfort, Price) according to their preferences. For example, a user who prioritizes safety over cost may assign a higher weight to the safety score. The system has a user interface, and the user can give a threshold value between 0 and 100 for each factor. If 100 is selected, the total score of the three-factor threshold will be 100. Since this will affect the score calculation, the order of the flights will also change. The algorithm has at least one default weight setting, which can be changed by the user for a personalized experience. The algorithm has predefined weights / thresholds for all factors, not just one factor. Users can change this threshold to represent which factor they attach more importance to. This will affect the ranking. Step 4: The composite score of each flight is calculated by multiplying the relevant weight of each normalized metric and adding these values. This score provides a single and combined assessment of the overall value of the flight based on safety, comfort, and price.

[0090] Step 5: Flights are sorted by their composite scores. Higher scores indicate the flights that best fit the weighted preferences of the user. The algorithm can also filter and recommend the flights according to certain user criteria, such as the preferred airlines, flight hours or transfer preferences.

[0091] Step 6: The algorithm is designed to process real-time data to ensure that the suggestions are always up-to-date and reflect the latest prices and available information.

[0092] In step 5, the order by the composite scores is as follows: After calculating the composite score from the normalized metric and thresholds, a composite score is obtained for each flight. A higher score means a better match. Then, the order is made and recommended according to those with higher scores.

[0093] The framework of the system, which initially focused on flight recommendations, can be scaled, and adapted for other travel-related applications, such as hotel or package tour recommendations. The algorithm can be adapted to a user-friendly interface that displays the flight options with their composite scores. It also provides tools for users to easily adjust the weights of different metrics.

[0094] In order for the invention to work effectively, each of the technical components that enable the method steps to work works together to create an advanced, versatile, and user-oriented flight recommendation system. This system comes to the forefront with its ability to balance a large number of critical travel factors, provide personalized recommendations and adapt to both user feedback and dynamic market conditions, and, as a result, improve the travel planning experience. Example of a Calculation:

[0095] Sample calculations to normalize the flight data using Min-Max normalization are as follows:

[0096] Normalized scores:

[0097] Flight A:

[0098] Price: $500 (Normalized: 0.0, lower is better)

[0099] Safety Score: 9.5 (Normalized: 1.0, higher is better)

[0100] Comfort Score: 7 (Normalized: 0.0, higher is better) Flight B:

[0101] Price: $750 (Normalized: 1.0)

[0102] Safety Score: 8.0 (Normalized: 0.0)

[0103] Comfort Score: 9 (Normalized: 1.0)

[0104] Flight C:

[0105] Price: $650 (Normalized: 0.6)

[0106] Safety Score: 9.0 (Normalized: 0.67)

[0107] Comfort Score: 8 (Normalized: 0.5)

[0108] Weights / threshold:

[0109] Price: 40% (0.4)

[0110] Safety Score: 40% (0.4)

[0111] Comfort Score: 20% (0.2)

[0112] Composite Score Calculations:

[0113] Flight A:

[0114] Composite Score=( 1 -0.0) * 0.4+1.0 * 0.4+0.0 *0.2

[0115] =1.0x0.4+1.0x0.4+0.0x0.2

[0116] = 0.4+0.4+0.0=0.8

[0117] Flight B:

[0118] Composite Score=( 1-1.0) x 0.4+0.0x0.4+1.0x0.2

[0119] =0.0x0.4+0.0x0.4+l.0x0.2

[0120] =0.0+0.0+0.2 =0.2

[0121] Flight C:

[0122] Composite Score=( 1-0.6) *0.4+0.67 * 0.4+0.5*0.2

[0123] =0.4*0.4+0.67*0.4+0.5*0.2

[0124] =0.16+0.268+0.1

[0125] =0.528

[0126] That is, the final composite scores are as follows:

[0127] Flight A: 0.8

[0128] Flight B: 0.2

[0129] Flight C: 0.528

[0130] These scores indicate that the most recommended flight by criteria and weights is flight A, followed by flight C, and the least recommended flight is flight B.

[0131] In summary, the system / algorithm has many advantages over other systems:

[0132] Use of multiple factors not used by other systems for recommendation.

[0133] Lack of problems of other systems such as cold start problem in factors, subjectivity of user information,

[0134] Factor Weights are customizable by the user for more personalized recommendations, as well as being pre-definable.

[0135] The terms or units mentioned in the system and method subject to the invention and their meanings:

[0136] Composite Scoring - Combining multiple elements into a single score.

[0137] Safety Score - The score for evaluating the safety information or records of the flight.

[0138] Comfort Score - A measure that evaluates in-flight experience or quality of experience.

[0139] Flight Price - It represents the cost factor of the flight.

[0140] Holistic and Comprehensive Evaluation - A term that emphasizes the comprehensive evaluation of flight options. User-Oriented Approach - It focuses on system design tailored to individual user needs. It focuses on the personalization of the user.

[0141] Personal Preferences and Priorities - It highlights the algorithm's ability to meet individual needs.

[0142] Customizable Flexible Tool - It highlights the versatility and customization capabilities of the system.

[0143] All in One Score - A unified measure that reflects the overall value and attractiveness of flights.

[0144] Application Format: The format usually includes:

[0145] - Software as a Service (SaaS): The algorithm is offered as a cloud-based service, allowing easy integration with existing platforms and systems.

[0146] - Mobile and Web Applications: For direct consumer use, the system can be used as a mobile application or web-based application that provides accessible and user-friendly interfaces.

[0147] - Application Programming Interface (API) Integration: The algorithm can be packaged as an API and allows other services to integrate this functionality into their own platforms.

[0148] - Customizable Control Panels: For corporate or airline use, the system can include customizable control panels for data analysis and decision-making.

[0149] Applicability of the Invention to the Industry

[0150] This invention is related to the multi-factor scoring algorithm for flight recommendations and allows the user to plan and book a trip more effectively, especially for use in the travel and tourism industry, and is highly applicable to the industry in this respect. The application format can be implemented in a variety of ways that appeal to both businesses and consumers. In general, the sectors it will serve are as follows: a. Integration into Online Travel Agencies (OTAs) and Booking Platforms: The algorithm can be incorporated into existing online booking platforms. These platforms can use the system to provide advanced flight search and recommendation features and to provide customers with more comprehensive and personalized options. b. Developing an Independent Travel Advice Application: The invention can form the basis of an independent application specially designed for flight recommendations. This application can be used by passengers to find the flights that best suit their preferences according to the algorithm's composite scoring system. c. Adoption by Airlines for Personalized Marketing: Airlines can use the algorithm to tailor their marketing efforts. By understanding customer preferences in terms of safety, comfort and price, airlines can offer targeted promotions and deals that are more likely to appeal to individual customers. d. Travel Management Tools for Companies: The system can be integrated into the travel management tools used by institutions. It can help businesses optimize their travel planning according to employees' preferences and corporate travel policies, and balance cost with comfort and safety. e. Inclusion in Travel Comparison and Review Websites: The algorithm can improve travel comparison and reviewing websites by adding a more complex method to rank and compare flight options and provide users with a clear, easy to understand measurement when making travel decisions. f. Travel Analytics and Market Research: The data collected and processed by the algorithm can be invaluable in terms of market research and analytics and can help industry players to understand consumer trends and preferences.

[0151] Given its versatility and adaptability, the invention has significant potential for widespread application in the travel industry, improving how flight information is processed, presented, and used for both individual passengers and industry players.

[0152] The invention is not limited to the above explanations; however, a person skilled in the art can easily present different embodiments of the invention. They must be assessed within the scope of the protection claimed by the claims of the invention.

Claims

CLAIMS1. A flight recommendation method (100) based on at least one factor, characterized in that it includes the following steps to provide an appropriate recommendation to the user in a short time by considering a large number of critical factors that affect the passenger's decisionmaking process at the same time:The user enters the number of flights and flight details through an interface (101) The user uses three predefined thresholds or enters at least three thresholds for each factor (safety, comfort, flight price) via the interface (102)Said threshold values are set predefined (i.e. automatic) by at least one data collection / processing / integration unit or the user enters the threshold values manually (103)Calculation of at least one final threshold value by the relevant data collection / processing / integration unit using the automatic threshold setting or manually entered threshold value (104)Calculation of at least one safety score metric, at least one accident rate, at least one audit and at least one death rate information by the data collection / processing / integration unit in question, using THE user-selected or predefined threshold value for the safety score (105)Calculation of at least one flight comfort score and at least one flight price metric using THE predefined threshold value for each factor by the relevant data collection / processing / integration unit (106)Calculation of a composite score by the relevant data collection / processing / integration unit by combining the calculated safety rate value, comfort rate value and flight metric value (107)Sorting of flights according to the composite score found (108)Displaying the output of the sorted lists to the user through an interface (109)2. A method according to claim 1, characterized in that at least three data mentioned in steps105 and 106, such as at least one safety score, at least one comfort score and at least oneflight price are collected in at least one database included in the system and integrated into the system.

3. A method according to claim 2, characterized in that the metrics mentioned in steps 105 and 106, such as safety score, comfort score and flight price, are normalized to a common scale with at least one data collection / processing / integration unit using a minimum-maximum normalization method.

4. A method according to claim 3, characterized in that customizable weighting is performed by adjusting the weight assigned to each metric according to users' preferences, such as safety score, comfort score and flight price as mentioned in steps 105 and 106.

5. A method according to claim 4, characterized in that a composite score in step 107 is calculated by multiplying each normalized metric by its respective weight and summing these values.

6. A method according to claim 5, characterized in that the composite scores of the flights are ranked according to the flights that best fit the weighted preferences of the user, and also in that the flights are filtered and recommended to the user according to specific user criteria, such as preferred airlines, flight times or transfer preferences.

7. A method according to claim 6, characterized in that at least one dynamic and real-time data processing unit is operated, which processes real-time data to ensure that the recommendations are always up to date, so that the latest prices and available information are reflected to the user via an interface.

8. A method according to claim 7, characterized by a safety score comprising a combination of at least one death score per airline, at least one accident score and at least one internationally recognized audit.

9. A method according to claim 8, characterized by a comfort score comprising at least one offered amenity and at least one other in-flight comfort parameter.

10. A method according to claim 9, characterized by at least one flight price including at least one price obtained from at least one airline database and at least one flight comparison website.

Images