update relevance filtering to allow non relevant only mode

README.md

@@ -134,11 +134,7 @@ The **control** language is English, and contains the queries for all 251 territ
 
 The dataset contains two types of relevance annotations:
 
-1. **Human Annotations**:
-
-   - Provided by three annotators for a subset of query-document pairs.
-   - Relevance is determined by majority vote across annotators.
-   - Territories are listed per annotator, capturing individual perspectives.
+1. **Human Annotations**: Provided by multiple annotators for a subset of query-document pairs and relevance is determined by majority vote across annotators.
 
 2. **LLM Annotations**:
    - Includes two modes:
@@ -227,10 +223,10 @@ ds_m3_zhs1 = load_dataset("borderlines/bordirlines", "zhs", split="m3.en")
 ds_m3_zhs2 = load_dataset("borderlines/bordirlines", "zhs", split="m3.qlang")
 
 # Load Dataset for English, relevant-only with human annotations
-ds_human_en = load_dataset("borderlines/bordirlines", "en", relevant_only=True, annotation_type="human")
+ds_human_en = load_dataset("borderlines/bordirlines", "en", relevance_filter="relevant", annotation_type="human")
 
-# Load Dataset for Simplified Chinese, few-shot LLM mode
-ds_llm_fewshot_zhs = load_dataset("borderlines/bordirlines", "zhs", relevant_only=True, annotation_type="llm", llm_mode="fewshot")
+# Load Dataset for Simplified Chinese, few-shot LLM mode, only non-relevant
+ds_llm_fewshot_zhs = load_dataset("borderlines/bordirlines", "zhs", relevance_filter="non-relevant", annotation_type="llm", llm_mode="fewshot")
 ```
 
 ## Citation

bordirlines.py

@@ -95,11 +95,11 @@ class BordIRLinesDataset(datasets.GeneratorBasedBuilder):
         for lang in SUPPORTED_LANGUAGES
     ]
 
-    def __init__(self, *args, relevant_only=False, annotation_type=None, llm_mode="fewshot", **kwargs):
+    def __init__(self, *args, relevance_filter="all", annotation_type=None, llm_mode="fewshot", **kwargs):
         super().__init__(*args, **kwargs)
-        self.relevant_only = relevant_only
+        self.relevance_filter = relevance_filter  # "relevant", "non-relevant", or "all"
         self.annotation_type = annotation_type
-        self.llm_mode = llm_mode  # Choose between "zeroshot" and "fewshot". Default: "fewshot".
+        self.llm_mode = llm_mode  # Default to "fewshot"
 
     def _info(self):
         return datasets.DatasetInfo(
@@ -116,7 +116,7 @@ class BordIRLinesDataset(datasets.GeneratorBasedBuilder):
                     "doc_text": datasets.Value("string"),
                     "doc_lang": datasets.Value("string"),
                     "relevant_human": datasets.Value("bool"),
-                    "territory_human": datasets.Sequence(datasets.Value("string")),
+                    "territory_human": datasets.Value("string"),
                     "relevant_llm_zeroshot": datasets.Value("bool"),
                     "relevant_llm_fewshot": datasets.Value("bool"),
                 }
@@ -195,27 +195,12 @@ class BordIRLinesDataset(datasets.GeneratorBasedBuilder):
 
             # Get Human Data
             human_data = human_map.get((query_id, doc_id), {})
-            # Parse relevant_human_votes manually
-            raw_votes = human_data.get("relevant_human", "[]")
-            relevant_human_votes = [
-                True if v.strip() == "True" else False if v.strip() == "False" else False
-                for v in raw_votes.strip("[]").split(",")
-                if v.strip()
-            ]
-
-            # Parse territory_human manually
-            raw_territories = human_data.get("territory_human", "[]")
-            territory_human = [
-                v.strip().strip("'").strip('"')  # Remove extra quotes and whitespace
-                for v in raw_territories.strip("[]").split(",")
-                if v.strip()
-            ]
-
-            # Calculate majority relevance
-            majority_relevant_human = (
-                sum(relevant_human_votes) > len(relevant_human_votes) / 2 if relevant_human_votes else False
-            )
 
+            # Directly use the new 'relevant' column (no need for tie-breaking)
+            relevant_human = human_data.get("relevant", False)  # Default to False if missing
+
+            # Directly use the 'territory' column instead of processing 'territory_human'
+            territory_human = human_data.get("territory", "")
 
             # Get LLM Data
             llm_data = llm_map.get((query_id, doc_id), {})
@@ -224,15 +209,25 @@ class BordIRLinesDataset(datasets.GeneratorBasedBuilder):
                 if self.llm_mode == "fewshot"
                 else llm_data.get("relevant_zeroshot", None)
             )
-            # Filtering logic
-            if self.relevant_only:
-                if self.annotation_type == "human" and not majority_relevant_human:
+
+            # Filtering logic based on relevance preference
+            if self.relevance_filter == "relevant":
+                if self.annotation_type == "human" and not relevant_human:
                     continue
                 elif self.annotation_type == "llm" and not (relevant_llm is True):
                     continue
-                elif not majority_relevant_human and not (relevant_llm is True):
+                elif not relevant_human and not (relevant_llm is True):
+                    continue
+
+            elif self.relevance_filter == "non-relevant":
+                if self.annotation_type == "human" and relevant_human:
+                    continue
+                elif self.annotation_type == "llm" and relevant_llm is True:
+                    continue
+                elif relevant_human or relevant_llm is True:
                     continue
 
+            # If "all", do not filter anything
 
             yield (
                 counter,
@@ -246,11 +241,10 @@ class BordIRLinesDataset(datasets.GeneratorBasedBuilder):
                     "doc_id": doc_id,
                     "doc_text": docs[doc_lang][doc_id],
                     "doc_lang": doc_lang,
-                    "relevant_human": majority_relevant_human,
+                    "relevant_human": relevant_human,
                     "territory_human": territory_human,
                     "relevant_llm_zeroshot": llm_data.get("relevant_zeroshot", None),
                     "relevant_llm_fewshot": llm_data.get("relevant_fewshot", None),
                 },
             )
-
-            counter += 1
+            counter+=1