update

README.md

@@ -39,7 +39,7 @@ limitations under the License.
 
 **News** :newspaper:
 
-- [ ] OAEI preparation and bug fixing. (**unreleased**)
+- [X] Deploy OAEI utilities at `deeponto.align.oaei` for scripts at the sub-repository [OAEI-Bio-ML](https://github.com/KRR-Oxford/OAEI-Bio-ML) as well as bug fixing. (**v0.8.4**)
 - [X] Bug fixing for BERTMap (stuck at reasoning) and ontology alignment evaluation. (**v0.8.3**)
 - [X] Deploy `deeponto.onto.OntologyNormaliser` and `deeponto.onto.OntologyProjector` (**v0.8.0**).
 - [X] Upload Java dependencies directly and remove mowl from pip dependencies (**v0.7.5**).

docs/bio-ml.md

@@ -25,7 +25,7 @@ $\textsf{Bio-ML}$ incorporates **five** ontology pairs for both equivalence and
 - **Subsumption Mapping Construction**: Reference subsumption mappings are constructed from the reference equivalence mappings, subject to target class deletion, i.e., if an equivalence mapping is used for constructing a subsumption mapping, its target ontology class will be removed to enforce direct subsumption matching (see [Ontology Pruning](#subsumption-mapping-construction)). 
 - **Candidate Mapping Generation**: To evaluate an OM system using ranking-based metrics, we generate a list of negative candidate mappings for each reference mapping using different heuristics (see [Ontology Pruning](#candidate-mapping-generation)).
 - **Locality Module Enrichment** (NEW :star2:): In the OAEI 2023 version, we enrich the pruned ontologies with classes that serve as context (and marked as **not used in alignment**) for the existing classes, using the **locality module** technique ([code](https://github.com/ernestojimenezruiz/logmap-matcher/blob/master/src/test/java/uk/ac/ox/krr/logmap2/test/oaei/CreateModulesForBioMLTrack.java)). OM systems can utilise these additional classes as auxiliary information while omitting them in the alignment process; the final evaluation will omit these additional classes as well.
-- **Special Sub-Track for Large Language Models** (NEW :star2:): In the OAEI 2023 version, we introduce a special sub-track for Large Language Model (LLM)-based OM systems by extracting small but challenging subsets from NCIT-DOID and SNOMED-FMA (Body) datasets. See [Special Sub-track for LLMs](#special-sub-track-for-llms) for detail.
+- **Bio-LLM: A Special Sub-Track for Large Language Models** (NEW :star2:): In the OAEI 2023 version, we introduce a special sub-track for Large Language Model (LLM)-based OM systems by extracting small but challenging subsets from NCIT-DOID and SNOMED-FMA (Body) datasets. See [Bio-LLM OAEI 2023](#bio-llm-oaei-2023) for detail.
 
 ## Links
 
@@ -180,7 +180,7 @@ Detailed structure is presented in the following figure (not yet available).
   <img alt="deeponto" src="../images/bio-ml-oaei-2023" height="420" style="width: 70%;">
 </p>
 
-## Special Sub-Track for LLMs
+## Bio-LLM OAEI 2023
 
 As Large Language Models (LLMs) are trending in the deep learning community, we formulate a special sub-track for LLM-based OM systems. For efficient and insightful evaluation, we select two small yet representative subsets from the NCIT-DOID and SNOMED-FMA (Body) datasets, each consisting of 50 **matched** and 50 **unmatched** class pairs. 
 

docs/changelog.md

@@ -7,9 +7,11 @@ Removed for now removed features.
 Fixed for any bug fixes.
 Security in case of vulnerabilities. -->
 
-## Unreleased
 
+## Unreleased 
 
+...
+## v0.8.4 (2023 July)
 ### Added
 
 - [X] **Add** specific check of `use_in_alignment` annotation in BERTMap for the OAEI.