Upload README.md with huggingface_hub

README.md

@@ -11,39 +11,6 @@ tags:
 - marked-temporal-point-process
 size_categories:
 - n<1K
-configs:
-- config_name: default
-  data_files:
-  - split: test
-    path: data/test-*
-dataset_info:
-  features:
-  - name: seq_idx
-    dtype: int64
-  - name: seq_len
-    dtype: int64
-  - name: type_category
-    dtype: string
-  - name: span_weeks
-    dtype: float64
-  - name: description
-    dtype: string
-  - name: metadata
-    dtype: string
-  - name: time_since_start
-    list: float64
-  - name: time_since_last_event
-    list: float64
-  - name: type_event
-    list: string
-  - name: type_text
-    list: string
-  splits:
-  - name: test
-    num_bytes: 8869501
-    num_examples: 311
-  download_size: 4326504
-  dataset_size: 8869501
 ---
 
 # Amazon Product Review Events
@@ -57,9 +24,9 @@ Sequences are aggressively filtered to prevent pattern exploitation — uninform
 - **Source:** [Amazon Reviews 2023](https://huggingface.co/datasets/McAuley-Lab/Amazon-Reviews-2023) (McAuley Lab, HuggingFace)
 - **Categories:** Electronics, Books, Home & Kitchen, Beauty & Personal Care
 - **Grouping:** Events grouped by user within a single product category
-- **Sequences:** 279
-- **Sequence length:** 50–78 events per sequence (mean: 58.7)
-- **Event types:** 40 sub-categories
+- **Sequences:** 311
+- **Sequence length:** 50–78 events per sequence (mean: 58.3)
+- **Event types:** 40 sub-categories (no "other")
 - **Time unit:** weeks
 
 ## Schema
@@ -77,19 +44,19 @@ Each record is a dictionary with 10 fields:
 | `time_since_start` | list[float] | Time since the first event (in weeks) |
 | `time_since_last_event` | list[float] | Time since the previous event (in weeks) |
 | `type_event` | list[str] | Product sub-category slug (see below) |
-| `type_text` | list[str] | Natural language description: product title, reviewer's star rating, review title, and review body. HTML tags, entities, embedded media tags, and URLs are stripped. |
+| `type_text` | list[str] | Natural language description with full category path (excluding root), product title, star rating, review title, and review body. HTML tags, entities, embedded media tags, and URLs are stripped. |
 
 ## Event Types (40 sub-categories)
 
-Event types are the 2nd-level product sub-categories, normalized to lowercase slugs. Rare sub-categories are mapped to `"other"` during initial curation, and all `"other"` events are then stripped from the final sequences.
+Event types are the 2nd-level product sub-categories, normalized to lowercase slugs. Rare sub-categories are mapped to `"other"` during initial curation, and all `"other"` events are then stripped from the final sequences. Products without a proper category hierarchy (e.g., generic "All Electronics") are excluded.
 
-**Home & Kitchen** (205 sequences, 10 types): `kitchen_dining` (2,719), `home_d_cor_products` (2,458), `bedding` (1,949), `furniture` (1,241), `storage_organization` (1,081), `bath` (1,079), `amazon_home` (539), `heating_cooling_air_quality` (317), `wall_art` (304), `seasonal_d_cor` (288)
+**Home & Kitchen** (211 sequences, 10 types): `kitchen_dining` (2,821), `home_d_cor_products` (2,504), `bedding` (2,033), `furniture` (1,343), `bath` (1,106), `storage_organization` (1,061), `heating_cooling_air_quality` (390), `wall_art` (345), `seasonal_d_cor` (304), `event_party_supplies` (239)
 
-**Beauty & Personal Care** (57 sequences, 10 types): `hair_care` (696), `skin_care` (693), `makeup` (642), `tools_accessories` (483), `foot_hand_nail_care` (437), `personal_care` (163), `shave_hair_removal` (127), `fragrance` (108), `salon_spa` (80), `men_s_grooming` (3)
+**Beauty & Personal Care** (76 sequences, 10 types): `hair_care` (928), `skin_care` (895), `makeup` (892), `tools_accessories` (656), `foot_hand_nail_care` (609), `personal_care` (206), `shave_hair_removal` (160), `fragrance` (144), `salon_spa` (90), `men_s_grooming` (10)
 
-**Electronics** (11 sequences, 10 types): `computers_accessories` (131), `television_video` (113), `home_audio` (98), `camera_photo` (73), `portable_audio_video` (58), `accessories_supplies` (43), `headphones_earbuds_accessories` (39), `security_surveillance` (38), `car_vehicle_electronics` (29), `all_electronics` (12)
+**Electronics** (18 sequences, 10 types): `computers_accessories` (224), `television_video` (183), `home_audio` (142), `camera_photo` (110), `portable_audio_video` (105), `headphones_earbuds_accessories` (79), `car_vehicle_electronics` (74), `accessories_supplies` (65), `security_surveillance` (59), `gps_finders_accessories` (9)
 
-**Books** (6 sequences, 10 types): `arts_photography` (55), `literature_fiction` (48), `history` (47), `books` (46), `children_s_books` (46), `biographies_memoirs` (29), `crafts_hobbies_home` (29), `mystery_thriller_suspense` (22), `politics_social_sciences` (7), `christian_books_bibles` (7)
+**Books** (6 sequences, 10 types): `literature_fiction` (82), `history` (56), `biographies_memoirs` (41), `children_s_books` (40), `mystery_thriller_suspense` (30), `teen_young_adult` (23), `arts_photography` (22), `christian_books_bibles` (18), `crafts_hobbies_home` (12), `politics_social_sciences` (8)
 
 ## Curation Filters
 
@@ -101,35 +68,37 @@ Sequences are selected to represent moderately prolific reviewers with diverse,
 | `max-events` | 80 | Max reviews per user sequence |
 | `min-types` | 3 | At least 3 distinct sub-category types |
 | `min-span` | 6 months | Exclude sequences spanning less than 6 months |
-| `max-span` | 60 months | Exclude sequences spanning 60+ months |
+| `max-span` | 72 months | Exclude sequences spanning 72+ months |
 | `max-sub-categories` | 10 | Top 10 sub-categories kept per parent category; rest mapped to `"other"` |
 | `min-subcat-count` | 50 | Sub-categories with fewer than 50 products mapped to `"other"` |
+| Verified purchase only | — | Only verified purchase reviews are included |
+| Exclude catch-all categories | — | Products without proper sub-category hierarchy (e.g., "All Electronics") are excluded |
 | Remove "other" events | — | All `"other"` events stripped; `time_since_last_event` recomputed |
 | `max-dominant-ratio` | 0.30 | Drop sequences where any single type exceeds 30% of events |
 | `max-repeat-ratio` | 0.35 | Drop sequences where consecutive same-type rate exceeds 35% |
 
 ## Event Text
 
-Each event's `type_text` contains product and review details without explicit sub-category labels:
+Each event's `type_text` is a natural language sentence containing the full category path (excluding the root category), product title, star rating, and review content:
 
-> The user reviewed a kitchen & dining product, "Instant Pot Duo 7-in-1 Electric Pressure Cooker", and gave it 5 out of 5 stars. Their review titled "Best kitchen purchase ever" says: This has completely changed how I cook dinner. The pressure cooking function is incredible and saves so much time.
+> The user reviewed "Instant Pot Duo 7-in-1 Electric Pressure Cooker" under Kitchen & Dining > Small Appliances > Multi-Cookers, rating it 5 out of 5 stars. Their review titled "Best kitchen purchase ever" says: This has completely changed how I cook dinner. The pressure cooking function is incredible and saves so much time.
 
-The model must learn to predict the product sub-category from review content, product titles, and rating patterns.
+The model must learn to predict the product sub-category from review content, product titles, category paths, and rating patterns.
 
 ## Example
 
 ```json
 {
   "seq_idx": 0,
-  "seq_len": 55,
+  "seq_len": 61,
   "type_category": "Electronics",
-  "span_weeks": 154.43,
-  "description": "Amazon Electronics review timeline for a user spanning Mar 2020 to Feb 2023. This sequence tracks the user's product review activity within the Electronics category on Amazon.",
-  "metadata": "{\"user_id\": \"AFXCEYAE...\", \"parent_category\": \"Electronics\", \"num_reviews\": 97, \"num_sub_categories\": 10, \"date_range_start\": \"2020-03-03T14:23:19.813000+00:00\", \"date_range_end\": \"2023-02-17T14:45:31.498000+00:00\", \"span_weeks\": 154.43, \"time_unit\": \"weeks\"}",
-  "time_since_start": [0.0, 0.58, 1.12, ...],
-  "time_since_last_event": [0.0, 0.58, 0.54, ...],
-  "type_event": ["home_audio", "computers_accessories", "portable_audio_video", ...],
-  "type_text": ["The user reviewed a home audio product, \"eoocvt 9.85ft T-Shape Car Mobile Radio DC Power Cable...\", and gave it 5 out of 5 stars. ...", ...]
+  "span_weeks": 269.34,
+  "description": "Amazon Electronics review timeline for a user spanning May 2015 to Jul 2020. This sequence tracks the user's product review activity within the Electronics category on Amazon.",
+  "metadata": "{\"user_id\": \"AF7S6K2L...\", \"parent_category\": \"Electronics\", \"num_reviews\": 76, \"num_sub_categories\": 10, ...}",
+  "time_since_start": [0.0, 0.0016, 0.0024, ...],
+  "time_since_last_event": [0.0, 0.0016, 0.0008, ...],
+  "type_event": ["camera_photo", "computers_accessories", "accessories_supplies", ...],
+  "type_text": ["The user reviewed \"D-Link DCS-5222L HD Pan & Tilt Wi-Fi Camera\" under Camera & Photo > Video Surveillance > Surveillance Cameras > Dome Cameras, rating it 5 out of 5 stars. ...", ...]
 }
 ```