公开(公告)号：MY193886A

公开(公告)日：2022-10-31

申请号：MYPI2017705130

申请日：2017-12-29

Applicant: MIMOS BERHAD

Inventor： KHALIL BOUZEKRI ,  MOHAMMAD ARSHI SALOOT ,  THENMALAR A/P ULANGANATHAN ,  ALI EBRAHIMI ,  ONG HONG HOE

IPC: G06F17/00

Abstract: The invention relates to a system (100) for normalizing context-free text generated by a user, comprising an out-of-vocabulary detection module (300) for detecting out-of- vocabulary terms from the text (102); one or more candidate generating modules (400, 500, 600, 700) which generates candidates for each of the out-of-vocabulary terms; a user profile builder (120) which retrieves user data and extracts user profile entities from the user data; and a candidate selection module (800); wherein the candidate selection module (800) identifies user profile entities that correspond to the candidates generated, examines the relationship between each of the user profile entities and the user to assign a probability score to each entity, and selects candidate based on the probability score for normalizing the terms in the text (102). The invention also relates to a method for normalizing text thereof.

公开(公告)号：MY181567A

公开(公告)日：2020-12-29

申请号：MYPI2015001695

申请日：2015-06-30

Applicant: MIMOS BERHAD

Inventor： DICKSON LUKOSE ,  NOR AZLINAYATI ABDUL MANAF ,  KHALIL BOUZEKRI

Abstract: A system (100) and method for semantic-based data harmonization is disclosed. The system comprises a reference table identifier (101), a table schema identifier (102) and a semantic code recommender (203) configured to recommend the relevant semantic code for the identified entities and assign the recommended semantic code for mapping. The method comprises determining (S201) whether each database table is a reference table, identifying (S203) the schema of the database table, extracting the entity of the database schema and content, and recommending (S204) the relevant semantic code for the identified entities from the database table. The method further comprises assigning the relevant semantic code to the identified entities for mapping and codifying (S210) the database schema and content with the relevant semantic codes. Figure 2

公开(公告)号：MY188277A

公开(公告)日：2021-11-25

申请号：MYPI2015704001

申请日：2015-11-05

Applicant: MIMOS BERHAD

Inventor： BENJAMIN CHU MIN XIAN ,  KHALIL BOUZEKRI ,  SHAZZAT HOSSAIN ,  TARIQUE AZIZ ,  DICKSON LUKOSE

IPC: G06F17/00

Abstract: The present invention relates to a system (100) and method for matching two or more documents. The system (100) for semantic matching of documents comprising a semantic parser (10) configured to parse the document to identify simple and complex attributes, a semantic matcher (30) configured to identify a plurality of candidate documents, filter out the candidates documents, and compute semantic similarity of documents using various semantic relationships between attributes, and a data sources manager (40) configured to manage input data sources from a data repository. The system (100) further comprising a hidden semantic knowledge extractor (20) configured to extract hidden semantic relationship between attributes and to create a discovered knowledge graph (56) from the hidden semantic relationship between attributes.

公开(公告)号：MY182742A

公开(公告)日：2021-02-04

申请号：MYUI2014703501

申请日：2014-11-25

Applicant: MIMOS BERHAD

Inventor： KHALIL BOUZEKRI ,  JASBEER SINGH A/L ATMA SINGH ,  FAROUQ HATEM HAMAD ,  DICKSON LUKOSE

Abstract: The present invention relates to a system for providing a learning plan (100). The system (100) is characterised by a Student Model repository (110) to store students? details, a Question Bank repository (120) to store questions, a Cluster Data Storage repository (130) to store students? cluster profile, a Learning Material repository (140) to store learning materials, an Assessment Based Learning Planner module (150) to assess and assign students with knowledge states and learning plans, a Cluster Based Learning Planner module (160) to create student profile clustering, a Personalized Learning Plan Selector module (170) to present students with learning materials based on the matched knowledge states and learning plans, a Knowledge Space Theory Library (180) to store functions for Knowledge Structure building and a Psych Library (190) to store functions and criteria for cluster profiling.

公开(公告)号：MY190232A

公开(公告)日：2022-04-06

申请号：MYPI2014703769

申请日：2014-12-12

Applicant: MIMOS BERHAD

Inventor： KHALIL BOUZEKRI ,  SHAZZAT HOSSAIN ,  DICKSON LUKOSE

Abstract: The present invention relates to a system (1000) and method for performing a semantic matching process by pruning search space right at the beginning of the semantic matching process. The system (1000) pre-processes binary conceptual structures to identify the most suitable starting point before it expands the starting point to get the best semantic matching between the two binary conceptual structures. The inputs of the system (1000) are two binary conceptual structures (G1, G2) and an ontology (800), while the final output is a percentage of an optimal semantic similarity between the two binary conceptual structures as well as the semantic matching that is used to compute the optimal semantic similarity. The most illustrative drawing: FIG. 1

公开(公告)号：MY176751A

公开(公告)日：2020-08-21

申请号：MYPI2015701631

申请日：2015-05-20

Applicant: MIMOS BERHAD

Inventor： KHALIL BOUZEKRI ,  SHAZZAT HOSSAIN ,  DICKSON LUKOSE

Abstract: The present invention relates to a system and method for diagnosing plant disease from an image. The system (1000) comprises a disease diagnosis component (100) having an image editor (110) to edit an input image. The system (1000) further comprising a disease management component (200) to select images based on priority ranking, extract image features of a plurality of system selected image (SSIMGs), compare user input image (UIIMG) with the SSIMGs, rank the SSIMGs based on priority ranking, notify the user with the SSIMGs and disease management information, and receive a user selected image (USIMG) and priority ranking and a plant data open repository (PDOR) (300) to store collection of knowledge bases of category of context and rearrange the plant disease images based on priority ranking. The disease diagnosis component (100) further includes a domain arbiter (120) to obtain necessary parameters, and contextualize the input to be processed. Figure 1

公开(公告)号：MY187722A

公开(公告)日：2021-10-14

申请号：MYPI2014002928

申请日：2014-10-14

Applicant: MIMOS BERHAD

Inventor： NOR AZLINAYATI ABDUL MANAF ,  NURUL AIDA OSMAN ,  KHALIL BOUZEKRI ,  DICKSON LUKOSE

Abstract: The present invention discloses a system (1) and method (10) for managing a collection of data (2) in table forms and for identifying reference tables based on multi-modality approaches. The system (1) and method (10) discloses a module (5) that is incorporated with a key filter to determine reference tables and eliminate non-reference tables as well as a ranking to rank reference tables in order to identify reference tables with minimal user intervention. Most illustrative drawing: Figure 1

公开(公告)号：MY177900A

公开(公告)日：2020-09-24

申请号：MYPI2015702232

申请日：2015-07-08

Applicant: MIMOS BERHAD

Inventor： BENJAMIN CHU MIN XIAN ,  LIU QIANG ,  KHALIL BOUZEKRI ,  DICKSON LUKOSE

Abstract: The present invention relates to a system (100) and method 5 for validating website usage. The system (100) comprising a content processor (10) to process a plurality of paragraphs of a website by using natural language processing. The system (100) further comprising a knowledge base aggregator (30) to harvest and index a plurality of knowledge bases from a Linked Data repository (50) to produce and maintain a domain mapping table, wherein the domain mapping table includes a plurality of knowledge base entries assigned with a trustworthiness confidence value (TCV), and a website validator (20) to validate intended usage of the website by utilising the domain mapping table. A method for validating website usage is characterised by the steps of processing contents of a website, harvesting a plurality of knowledge bases to produce and maintain a domain mapping table, and validating intended usage of the website by utilising a domain mapping table. Fig. 1