CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application is a continuation-in-part of U.S. Provisional Patent Application No. 63/974,577, filed February 3, 2026, titled "INTELLIGENT LENDER MATCHING SYSTEM WITH DYNAMIC PERFORMANCE SCORING AND ADAPTIVE QUALIFICATION ENGINE," the entire contents of which are incorporated herein by reference.

FIELD OF THE INVENTION

[0002] This continuation-in-part relates to improvements in financial technology systems for intelligent lender matching, specifically novel methods for calculating estimated loan terms, interest rates, approval likelihood scores, and generating personalized loan offers based on applicant-lender compatibility analysis.

BACKGROUND

[0003] The parent application (63/974,577) disclosed a system for matching loan applicants with lenders using machine learning approval probability prediction and performance-based lender scoring. However, the parent application did not disclose specific methodologies for:

• Calculating estimated interest rates based on applicant risk factors
• Determining optimal loan terms dynamically
• Generating weighted approval likelihood scores using non-ML algorithms
• Creating personalized loan offers with industry-specific messaging
• Displaying monthly payment estimates in real-time during matching

[0004] Existing loan comparison platforms show generic interest rate ranges (e.g., "7-15%") without personalization. Borrowers cannot see estimated monthly payments until after applying. There exists a need for real-time, personalized loan term estimation that provides actionable financial information during the matching phase.

SUMMARY OF NEW INNOVATIONS

[0005] This continuation-in-part adds the following novel systems and methods to the parent application:

Innovation 1: Dynamic Interest Rate Estimation Algorithm

[0006] A method for calculating personalized interest rate estimates comprising:

• Retrieving lender's base interest rate range
• Calculating credit score differential adjustment: creditAdj = f(applicant_score - lender_min_score)
• Calculating time-in-business adjustment: timeAdj = g(months_in_business - lender_min_months)
• Calculating loan-to-revenue ratio adjustment: sizeAdj = h(loan_amount / annual_revenue)
• Computing final estimated rate: estimatedRate = baseRate + creditAdj + timeAdj + sizeAdj
• Providing rate range: [estimatedRate - 0.5%, estimatedRate + 1.0%]

Innovation 2: Weighted Approval Likelihood Scoring

[0007] A non-machine-learning approval scoring method comprising:

• Credit score factor (40 points maximum): scored based on differential above minimum (10-40 points)
• Time in business factor (30 points maximum): scored based on months above minimum (5-30 points)
• Revenue factor (30 points maximum): scored based on revenue ratio above minimum (10-30 points)
• Total approval likelihood score: sum of three factors (0-100 scale)

Innovation 3: Adaptive Loan Term Determination

[0008] A rule-based system for determining optimal loan terms comprising:

• If loan_amount < $50,000 → term = 36 months
• If $50,000 ≤ loan_amount ≤ $250,000 → term = 60 months
• If loan_amount > $250,000 → term = 84 months
• If lender_type = online_lender OR merchant_cash_advance → term = min(term, 36)
• Calculating decision timeline based on lender's avg_decision_time_days
• Generating funding speed classification (Fast/Moderate/Standard)

Innovation 4: Real-Time Monthly Payment Calculation

[0009] A method for displaying monthly payment estimates during matching comprising:

• Computing monthly interest rate: monthlyRate = annualRate / 12 / 100
• Calculating monthly payment using standard amortization formula:
• payment = principal × (monthlyRate × (1 + monthlyRate)^months) / ((1 + monthlyRate)^months - 1)
• Displaying payment alongside approval probability in user interface

Innovation 5: Industry-Matched Personalized Offer Generation

[0010] A system for creating contextual loan offers comprising:

• Detecting industry match: comparing applicant's industry to lender's preferred_industries array
• Generating headline: if match → "Lender X specializes in [industry] businesses like yours"
• Extracting key benefits: unsecured loans, fast funding, industry expertise
• Creating offer structure with headline, benefits array, and available products

DETAILED DESCRIPTION OF NEW INNOVATIONS

Dynamic Interest Rate Estimation Algorithm - Technical Implementation

[0011] The interest rate estimation algorithm operates as follows:

[0012] This algorithm provides personalized rate estimates that reflect actual risk factors considered by lenders, improving upon generic rate ranges.

Weighted Approval Likelihood Scoring - Mathematical Formula

[0013] The approval likelihood score is calculated using three weighted factors:

[0014] The 40/30/30 weighting reflects industry standards where credit score is the primary factor, followed by business tenure and revenue capacity equally weighted.

Enhanced Matching Display Integration

[0015] The system integrates all new components into a unified display comprising:

• Lender ranking by approval_likelihood × lender_performance_score (from parent application)
• Personalized offer card showing industry-matched headline and key benefits
• Four-metric display: Estimated Rate | Loan Term | Monthly Payment | Decision Time
• Funding terms section: Collateral requirement | Funding speed | Max loan amount
• "Best Match" badge for highest-ranked lender

NOVEL ASPECTS AND ADVANTAGES

[0016] The new innovations provide several novel advantages over prior art:

Transparency Through Personalization:

Unlike generic rate ranges, applicants see personalized estimates based on their specific risk profile before applying.

Actionable Financial Information:

Monthly payment display allows immediate affordability assessment without requiring application submission.

Multi-Factor Risk Assessment:

The 40/30/30 weighted scoring provides deterministic approval likelihood independent of ML model availability.

Context-Aware Messaging:

Industry-matched offers improve relevance and trust through specialized positioning.

Adaptive Term Optimization:

Loan terms automatically adjust based on amount and lender characteristics for optimal matching.

CLAIMS

Claim 1.

A method for estimating personalized loan interest rates in a lender matching system, the method comprising:

• a) receiving a lender's base interest rate and stated minimum requirements;
• b) receiving an applicant's credit score, time in business, annual revenue, and requested loan amount;
• c) calculating a credit score differential adjustment based on the difference between the applicant's credit score and the lender's minimum credit score requirement;
• d) calculating a time-in-business adjustment based on the difference between the applicant's time in business and the lender's minimum time requirement;
• e) calculating a loan-to-revenue ratio adjustment based on the ratio of requested loan amount to annual revenue;
• f) computing an estimated interest rate by summing the base rate with the credit, time, and ratio adjustments;
• g) generating an estimated rate range by applying a variance to the estimated rate; and
• h) displaying the estimated rate to the applicant during the lender matching process.

Claim 2.

The method of claim 1, wherein the credit score differential adjustment is calculated as:

• 3.0 percentage points if the applicant's credit score is below the lender's minimum;
• 1.5 percentage points if the differential is between 0 and 50 points;
• 0.5 percentage points if the differential is between 50 and 100 points; and
• 0 percentage points if the differential exceeds 100 points.

Claim 3.

A system for calculating approval likelihood scores using weighted factors, the system comprising:

• a) a credit score analysis module that assigns 10-40 points based on applicant credit score differential above lender minimum, weighted at 40% of total score;
• b) a time-in-business analysis module that assigns 5-30 points based on months in business above lender minimum, weighted at 30% of total score;
• c) a revenue analysis module that assigns 10-30 points based on annual revenue ratio above lender minimum, weighted at 30% of total score;
• d) a scoring aggregation module that sums the weighted scores to produce a total approval likelihood score on a 0-100 scale; and
• e) a display module that presents the approval likelihood score alongside lender information.

Claim 4.

The system of claim 3, wherein the credit score analysis module assigns points according to the following thresholds:

• 40 points when credit differential ≥ 100 points;
• 30 points when credit differential is 50-99 points;
• 20 points when credit differential is 20-49 points; and
• 10 points when credit differential is 0-19 points.

Claim 5.

A method for determining optimal loan terms in an automated lending platform comprising:

• a) receiving a requested loan amount from an applicant;
• b) identifying a lender type from a database of lender characteristics;
• c) applying amount-based term rules: assigning 36 months if loan amount is less than $50,000; assigning 60 months if loan amount is between $50,000 and $250,000; and assigning 84 months if loan amount exceeds $250,000;
• d) applying lender-type modifications: reducing term to maximum 36 months if lender type is online_lender or merchant_cash_advance;
• e) retrieving lender's average decision time from historical data;
• f) classifying funding speed as Fast (<7 days), Moderate (7-14 days), or Standard (>14 days); and
• g) displaying the determined term, decision time, and funding speed classification to the applicant.

Claim 6.

A method for generating personalized loan offers in a lender matching interface comprising:

• a) identifying an applicant's industry sector;
• b) retrieving a lender's preferred industries from a lender profile;
• c) determining an industry match by comparing applicant industry to preferred industries;
• d) if match exists, generating a headline comprising: "[Lender Name] specializes in [industry sector] businesses like yours";
• e) if no match exists, generating a generic headline comprising: "[Lender Name] is ready to fund your business";
• f) extracting key benefits from lender characteristics including: no collateral requirement if offers_unsecured is true; fast funding if avg_decision_time < 7 days; and industry expertise if industry match exists;
• g) compiling offer structure comprising headline, benefits array, and available loan products; and
• h) displaying personalized offer in a highlighted interface element.

Claim 7.

A system for displaying real-time monthly payment estimates during loan matching comprising:

• a) retrieving an estimated interest rate from a rate calculation module;
• b) retrieving a determined loan term from a term determination module;
• c) receiving a requested loan amount from the applicant;
• d) calculating monthly interest rate as: annualRate / 100 / 12;
• e) computing monthly payment using the formula: payment = principal × (monthlyRate × (1 + monthlyRate)^months) / ((1 + monthlyRate)^months - 1);
• f) rounding the computed payment to the nearest dollar; and
• g) displaying the monthly payment estimate alongside the estimated rate, term, and approval likelihood in a unified lender information card.

Claim 8.

The method of claim 1, further comprising ranking multiple lenders by a composite score calculated as: (approval_likelihood) × (lender_performance_score from parent application) × (lender_priority_weight), wherein lenders are displayed in descending order of composite score.

Claim 9.

The system of claim 3, wherein the approval likelihood score is used in combination with the machine learning approval probability from the parent application to provide dual scoring methodologies.

Claim 10.

A comprehensive loan matching interface integrating innovations from claims 1, 3, 5, 6, and 7, the interface comprising:

• a ranked list of lenders sorted by composite matching score;
• a personalized offer card displaying industry-matched headlines and key benefits;
• a four-metric display panel showing estimated rate, loan term, monthly payment, and decision time;
• a funding terms section showing collateral requirements, funding speed, and maximum loan amount;
• a "Best Match" indicator for the highest-ranked lender; and
• an "Apply Now" button linking to the lender's application process.

INTEGRATION WITH PARENT APPLICATION

[0017] The innovations disclosed in this CIP work in conjunction with the parent application's core technologies:

• The interest rate estimation (Claim 1) enhances the ML approval probability by providing rate transparency
• The weighted scoring (Claim 3) provides a backup scoring methodology when ML model is unavailable
• The personalized offers (Claim 6) leverage the performance scoring system from parent to highlight reliable lenders
• All innovations maintain the parent's qualified rejection tracking and adaptive visibility mechanisms

ABSTRACT

[0018] A continuation-in-part adding novel methods for personalized loan term estimation to an intelligent lender matching platform. The system calculates estimated interest rates using multi-factor risk adjustments (credit, time, revenue ratio), determines approval likelihood using weighted scoring (40/30/30), adaptively assigns loan terms based on amount and lender type, generates industry-matched personalized offers, and displays real-time monthly payment estimates. These innovations provide borrowers with actionable financial information during the matching phase, improving transparency and decision-making capability compared to generic rate ranges in prior art.