updated readme

README.md

@@ -9,3 +9,75 @@ short_description: Predicts JIRA Story Point positional Increments
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+
+
+# **JIRA Story Point Increment Predictor**
+
+T
+The model uses JIRA **summary** and **description** text fields, converted into embeddings using the Hugging Face model  
+[`sentence-transformers/all-MiniLM-L6-v2`](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2).  
+
+Using **Databricks AutoML**, this dataset was analyzed and an **XGBoostRegressor** model was generated to predict story point increments.
+
+---
+
+## **Why**
+
+For any JIRA project facing time constraints in sizing issues, this solution provides a **generalized prediction strategy** using a **complexity scale of 8 increments**.  
+
+While mapping story points to hours is up to each team, this model offers a way to assign **complexity** using a **positional index** on that scale.
+
+---
+
+## **Integration**
+
+Hosting this model as an **API** allows for seamless integration with client services.  
+
+A client can send issue **summary** and **description** text, which the API embeds and passes to the model.  
+The model returns a **predicted increment** for story points.  
+
+Clients can then map this increment to their internal story point scale as desired.
+
+---
+
+## **Attribution**
+
+This model was built using data provided by:
+
+### **The Public Jira Dataset**
+
+**Creators**  
+- Montgomery, Lloyd  
+- Lüders, Clara  
+- Maalej, Prof. Dr. Walid  
+
+**Description**  
+Jira is an issue tracking system that helps software companies (among others) manage their projects, communities, and processes.  
+This dataset is a collection of **public Jira repositories** downloaded using the Jira API V2.  
+
+It includes data from:
+- **16 public Jira repositories**
+- **1822 projects**
+- **2.7 million issues**
+- **32 million changes**
+- **9 million comments**
+- **1 million issue links**
+
+The repository contains:
+- MongoDB data dumps  
+- Scripts to download and interpret the data  
+- Qualitative analyses to make the data more approachable  
+
+> **Note:**  
+> All personal information (e.g., assignee, creator, reporter, comment authors) has been anonymized using UUID4 masks, maintaining uniqueness while protecting privacy.
+
+---
+
+### **Citation**
+
+Montgomery L, Lüders C, Maalej W.  
+*An Alternative Issue Tracking Dataset of Public Jira Repositories.*  
+In: 2022 IEEE/ACM 19th International Conference on Mining Software Repositories (MSR).  
+2022 May 23. pp. 73–77. IEEE.  
+
+🔗 [https://zenodo.org/records/15719919](https://zenodo.org/records/15719919)