Commodities Industries Intelligence
收藏资源简介：
Bitvore's Commodities Industries Intelligence offers in-depth insights into the world of commodities industries such as raw materials, chemicals, and construction materials. Navigate the market with data on supply chain dynamics, production rates, and materials innovations. Key Benefits: - Informed Decisions: The latest intelligence on Commodities companies, market shifts, sustainability trends, emerging hotspots, potential risk zones, regulatory changes, industry sentiments and much more. - Deep Dive Analysis: Comprehensive coverage of 168 Finance Topics and 37 unique ESG Topics, all filtered to spotlight the Commodities sector. - Broad Data Coverage: Monitor 60k+ quality global sources, using publicly available and premium licensed data, tracking developments related to over 150,000 global Commodities companies, both public and private. Bitvore powers data-driven actionable signals through advanced, trusted and proven AI & ML processes, enabling our customers to manage risks and identify opportunities. Our NLP and ML models analyze unstructured content and are trained to identify 17 Finance Topics (with 151 Sub Topics) and 37 unique ESG material topics (tied to SASB, the Big 4 and other taxonomies). 60k+ quality globally sources monitored for 400k+ companies globally, using publicly available and premium licensed data. Bitvore delivers timely data, insights, signals and indicators of significant developments affecting companies, industries, bonds, markets and emerging themes including Bitvore Risk, Growth, ESG, E, S and G entity and article level sentiment scores. Some sample tables are included below: - ORGS: Contains data on organizations, which represent corporate entities, companies or organizations both public and private - RECORD: a table of articles about corporate entities, companies or organizations both public and private. - SIGNALS: Contains data about topics that are identified for any given company Some Sample fields from the RECORD table are included below: - ARTICLETYPE: Type of the record, News or Press Release. - SENTIMENT: Overall Sentiment of the record within a range of -1.0 (negative) and 1.0 (positive). - SOURCENAME: Name (typically hostname) of the source of the record. - ESGSENTIMENT: ESG Sentiment of the record within a range of -1.0 (negative) and 1.0 (positive). - SOURCEURL: URL for the source of the record. - KEY: 56 character unique identifier of the data record, a record may be on multiple rows if more than one company is associated with it. - PUBLISHED_DATE: Date/time the record was initially published at as yyyy-MM-dd hh:mm. - BODY: Body of any specified article - TITLE: Title of any specified article Some Sample fields from the ORGS table are included below: - ID: Bitvore ID that uniquely identifies the company - STATE: State code of the state the company’s headquarters is located in - EMPLOYEES: Number of employees for the company within a range. - TICKER: An abbreviation used to uniquely identify publicly traded companies - COUNTRY: Country the company’s headquarters is located in - LASTMODIFIED: The date a record was last changed as yyy-MM-dd hh:mm. It’s not often but on occasion a record can be changed to reflect either new information extracted from it or improvements to models to get more accurate information, such as a sentiment value. Bitvore Commodities Intelligence Data Sets also include “Layer 2” advanced tags for industries, people, keywords/phrases, relationships, and geography providing programmatic access to complex values and relationships. Industries – 3 layered hierarchy of Economic Sector > Industry Group > Business Sector People – 2.55 million plus database of reference data with accurate extraction and disambiguation Keywords/Phrases – Broad lexicon of useful text tied to conceptual meaning, scored by confidence and salience Relationships – Over 17 different types of NLP-derived relationships like Person-Title-Company or Subsidiary-of with scored evidence Geography – Bottom-up tagging of geographical references including named places and latitude/longitude mapping Sentiment – Per-entity and per use sentiment scoring specific to references/co-references Sentiment Scores - Bitvore Risk, Growth, ESG, E, S and G entity and article level sentiment scores.
Bitvore的大宗商品行业情报（Commodities Industries Intelligence）可为原材料、化工品、建筑材料等大宗商品领域提供深度行业洞察。依托供应链动态、生产速率及材料创新相关数据，助力用户精准研判市场走势。 核心优势： - 辅助科学决策：覆盖大宗商品企业最新情报、市场动向、可持续发展趋势、新兴热点区域、潜在风险地带、监管政策变动、行业情绪等多维度信息，助力用户做出明智决策。 - 深度剖析分析：涵盖168个金融主题（"Finance Topics"）与37个独特ESG主题（"ESG Topics"），所有内容均经过筛选，聚焦大宗商品赛道。 - 全面数据覆盖：追踪6万余个优质全球数据源，整合公开可用数据与授权付费数据，覆盖超过15万家全球大宗商品企业（含上市与非上市主体）的发展动态。 Bitvore依托先进、可靠且经过验证的人工智能与机器学习（"AI & ML"）流程，打造可落地的数据驱动型信号，助力客户管控风险、发掘机遇。其自然语言处理（"NLP"）与机器学习模型可对非结构化内容进行分析，经过训练后能够识别17个金融主题（含151个子主题）以及37个与可持续会计准则委员会（"SASB"）、四大会计师事务所（"Big 4"）及其他分类体系相关的实质性ESG主题。平台通过公开可用数据与授权付费数据，对全球6万余个优质数据源进行监测，覆盖全球超40万家企业。Bitvore可及时提供影响企业、行业、债券、市场及新兴主题的重大发展相关数据、洞察、信号与指标，其中包括Bitvore风险、增长、ESG、环境（"E"）、社会（"S"）及治理（"G"）维度的实体级与文章级情绪评分。 以下包含部分示例数据表： - ORGS：存储企业组织相关数据，涵盖上市与非上市的法人实体、公司及各类组织机构。 - RECORD：收录关于各类上市及非上市企业、组织机构的文章数据表格。 - SIGNALS：存储针对任意指定企业识别出的主题相关数据。 以下为RECORD表的部分示例字段： - ARTICLETYPE：记录类型，分为新闻（"News"）或新闻稿（"Press Release"）。 - SENTIMENT：记录整体情绪评分，取值范围为-1.0（负面）至1.0（正面）。 - SOURCENAME：记录来源的名称（通常为域名）。 - ESGSENTIMENT：记录的ESG情绪评分，取值范围为-1.0（负面）至1.0（正面）。 - SOURCEURL：记录来源的网址。 - KEY：数据记录的56位唯一标识符，若一条记录关联多家企业，则可能对应多行数据。 - PUBLISHED_DATE：记录首次发布的日期与时间，格式为yyyy-MM-dd hh:mm。 - BODY：对应文章的正文内容。 - TITLE：对应文章的标题。 以下为ORGS表的部分示例字段： - ID：用于唯一标识企业的Bitvore ID。 - STATE：企业总部所在地的州/省代码。 - EMPLOYEES：企业员工人数区间范围。 - TICKER：用于唯一标识上市交易企业的缩写代码。 - COUNTRY：企业总部所在国家。 - LASTMODIFIED：记录最后一次修改的日期与时间，格式为yyyy-MM-dd hh:mm。尽管此类情况并不常见，但偶尔会因从数据源中提取到新信息，或通过优化模型以获取更精准的信息（如情绪评分）而修改记录。 Bitvore大宗商品情报数据集还包含针对行业、人物、关键词/短语、关联关系及地理信息的"第二层（Layer 2）"高级标签，可通过程序化方式访问复杂的数值与关联关系： - 行业：采用三层层级结构，即经济部门>行业组别>业务板块。 - 人物：拥有超255万条参考数据的数据库，可实现精准的信息提取与歧义消解。 - 关键词/短语：涵盖与概念含义相关的广泛词汇库，可按置信度与显著性进行评分。 - 关联关系：包含超过17种由自然语言处理衍生的关联关系类型，例如“人物-职位-公司”或“子公司关系”，并附带评分佐证依据。 - 地理信息：自下而上对地理参考信息进行标注，包括命名地点及经纬度映射。 - 情绪评分：针对特定引用/共指关系的实体级与使用场景情绪评分。 - 情绪评分细则：Bitvore风险、增长、ESG、环境（"E"）、社会（"S"）及治理（"G"）维度的实体级与文章级情绪评分。
